p53 Induces skin aging by depleting Blimp1+ sebaceous gland cells

Abstract

p53 is an important inducer of organismal aging. However, its roles in the aging of skin remain unclear. Here we show that mice with chronic activation of p53 develop an aging phenotype in the skin associated with a reduction of subcutaneous fat and loss of sebaceous gland (SG). The reduction in the fat layer may result from the decrease of mammalian TOR complex 1 (mTORC1) activity accompanied by elevated expression of energy expenditure genes, and possibly as compensatory effects, leading to the elevation of peroxisome proliferator-activated receptor (PPAR)γ, an inducer of sebocyte differentiation. In addition, Blimp1(+) sebocytes become depleted concomitantly with an increase in cellular senescence, which can be reversed by PPARγ antagonist (BADGE) treatment. Therefore, our results indicate that p53-mediated aging of the skin involves not only thinning through the loss of subdermal fat, but also xerosis or drying of the skin through declining sebaceous gland activity.

Figure 1

Skin of p53^TSD/− mice reveals senescence with p53 activation but not apoptosis. (a) Skin of p53^TSD/− mice display hyperpigmentation. Macroscopic appearance of tails and foot pads of 4-week-old p53^+/− and p53^TSD/− mice. (b) The relative expression of genes related to apoptosis such as Noxa, Puma, and Bim, as well as genes related to senescence in the skin of p53^+/− and p53^TSD/− mice was determined by quantitative RT-PCR analysis. The mRNA levels of the senescence-related genes p21, p27, Pai1, and Tert1 are increased in the skin of p53^TSD/− mice, compared with those in p53^+/− mice. n=3. Values represent means±S.D. (c) The apoptotic cells were identified by TUNEL assay. Number of apoptotic cells per hair follicle in 4-week-old p53^+/− and p53^TSD/– mice was counted. n=30–35 HF for each genotype. Values represent means±S.D. (Magnification, × 50). (d) SA-β-gal staining in the skin of p53^+/− and p53^TSD/− mice (Magnification, × 50). (e) Quantification of hair follicles showing SA-β-gal activity and DNA damage foci and in the skin of p53^+/− and p53^TSD/− mice. Epidermal sheets of p53^+/− and p53^TSD/− mice were labeled for γ-H2A.X (red) as a senescent marker by immunofluorescence (IF) and representative images of sebaceous gland (SG) are shown. Quantification (the % of cells with DNA damage foci or SA-β-gal activity within SG) was shown in each panel. n=3–4 fields from two mice for each genotype. Values represent means±S.D. (Magnification, × 100)

Figure 2

Skin of p53^TSD/− mice exhibits the reduction of subcutaneous adipose tissue. (a) Hematoxylin-eosin (H&E) analyses of the skin of the p53^TSD/− and p53^+/− mice at P28 reveal the reduced subcutaneous adipose tissue. A, adipose tissue; E, epidermis; D, dermis. Magnification, × 50. (b) The thickness of dermis, epidermis, and subcutaneous adipose tissue of H&E-stained skin sections of the p53^TSD/− and control p53^+/− mice was measured with the measurement tool in ImageJ. Mean values from four mice for each genotype were presented with S.D. t-test was used to calculate the statistical significance. (c) Size of the subcutaneous adipocytes in H&E-stained skin sections of the p53^TSD/− and control p53^+/− mice was determined with the measurement tool in ImageJ. Mean values from four mice (>100 cells per mouse) for each genotype were presented with S.D. *P<0.05. (d) Food consumption of p53^TSD/−mice and p53^+/−mice. Three-week-old mice were monitored for 1 week. n=3. (e) Reduced adipose tissue in the skin of p53^TSD/− mice grafted onto the back of the same SCID mouse. The skin of p53^TSD/− and control p53^+/− mice were grafted onto the back of the same SCID mouse for four weeks and analyzed by H&E staining. The skin grafts were derived from newborn mice. The upper panel shows the image of skin grafts on the back of SCID mice. The middle and bottom images show the H&E-stained sections of the skin grafts in the same scale

Figure 3

The reduction of subcutaneous adipose tissue in skin of p53^TSD/− mice is correlated to the reduced mTORC1 activity and increased energy expenditure. (a) Phosphorylation of 4EBP1 in the skin of p53^TSD/− mice was reduced when compared with that in p53^+/− mice as determined by immunohistochemistry (IHC). (Magnification, × 50). (b) The phosphorylation of 4EBP1 and S6K was reduced in the skin of p53^TSD/− mice when compared with that in p53^+/− mice. Samples from epidermis and dermis/subcutaneous fat layer were analyzed by western blotting. (c, d) The relative expression of genes related to energy expenditure in the skin of p53^+/− and p53^TSD/− mice as determined by quantitative RT-PCR analysis. Values represent means±S.D. n=3. (e, f) The relative expression of PGC-1α and Ucp1 in the skin of p53^+/− and p53^TSD/− mice as determined by quantitative RT-PCR analysis. Samples from epidermis and dermis/subcutaneous fat layer were analyzed. Values represent means±S.D. n=3. (g) The thickness of the subcutaneous fat layer in the skin of 5-week-old p53^TSD/−Puma^−/− mice treated with or without Rapa for 7 days. The data were derived from skin sections stained with H&E. (h) The thickness of the subcutaneous fat layer in the skin of 4-week-old p53^TSD/−Puma^−/− mice treated with or without CSA for 7 days. (i) The thickness of the subcutaneous fat layer of 4-week-old p53^+/− mice treated with or without DNP for 7 days. (j) The thickness of the subcutaneous fat layer in the skin of starved and fed p53^+/− mice. (k) Reduced phosphorylation of 4EBP1 in the skin of starved p53^+/− mice. Phosphorylation of 4EBP (red) and CD34 (green) in the skin of starved and fed p53^+/− mice were shown by IF (Magnification, × 100)

Figure 4

Hyperactivation of p53 in the skin slows hair cycle, but does not impact on either hair re-growth after shaving or closure after wounding. (a) Macroscopic appearance of tails and footpads of 4-week-old p53^+/−Puma^−/− mice and p53^TSD/−Puma^−/− mice. p53^TSD/−Puma^−/− mice display hyperpigmentation. (b) The relative expression level of p21 in the skin of p53^+/−Puma^−/− mice and p53^TSD/−Puma^−/− mice was determined by quantitative RT-PCR analysis. n=3. Values represent means±S.D. Wound heal assay (c) and hair re-growth assay (d) in the skin of 2-month-old p53^+/−Puma^−/− mice and p53^TSD/−Puma^−/− mice. Values represent means±S.D. (e) H&E analyses in the skin of the p53^TSD/− Puma^−/− mice or p53^+/−Puma^−/− mice at P21, P24, P28, and P35. Data indicate that hair follicle cycle is delayed in p53-activated mouse

Figure 5

The induction of PPARγ in p53^TSD/− mice leads to the depletion of Blimp1⁺ progenitor of sebocyte. (a) The relative expression level of genes involved in adipogenesis in the skin of p53^+/− and p53^TSD/− mice determined by quantitative RT-PCR analysis. n=3. Values represent means±S.D. (b) IHC for Blimp1 in the skin of p53^TSD/− and p53^+/− mice. (Magnification, × 25). (c) Epidermal sheets of p53^+/−, p53^TSD/−, p53^TSD/−Puma^−/− mice, and p53^+/−Puma^−/− mice with or without PPARγ antagonist BADGE were labeled for Blimp1 (red) as a sebocyte progenitor marker by IF (Magnification, × 50). (d) FACS analysis of Blimp1 cells in the skin of p53^TSD/− mice. n=3 from three mice for each genotype. Values represent means±S.D. Cells expressing SG progenitor marker, Blimp1 reduced in the skin of p53^TSD/− mice (e) H&E analyses in the skin of the p53^TSD/−Puma^−/− mice and p53^+/−Puma^−/− mice at P24, P90, and P180. Data show that SG is rapidly depleted in p53^TSD/− mice and this atrophy is correlated to the reduction of subcutaneous fat layer. Arrow, SG (Magnification, × 100). (f) IHC for Blimp1 in the skin of P180-old p53^TSD/−Puma^−/− mice and p53^+/−Puma^−/− mice. Data show that Blimp1 cells are located in SG of p53^+/− mice, but depleted in p53^TSD/− mice (Magnification, × 25). (g–i) Reduced size and cell number of SG in p53^TSD/− mice. Epidermal sheets of p53^+/− and p53^TSD/− mice were labeled with lipid dye Nile Red (green) and DAPI. After taking pictures of IF in the same scale, the area of SG was measured and cells in this area were counted with ImageJ; n equal over 100 cells from three mice for each genotype. Values represent means±S.D. and t-test used to calculate the statistical significance, *P<0.05. (Magnification, × 100)

Figure 6

p53-induced senescence is related to PPARγ-dependent differentiation and ROS. (a) γ-H2A.X (red) in the epidermal sheets of p53^+/− mice and p53^TSD/− mice treated with or without BADGE is analysed by IF and representative images are shown. Total cells with DNA damage foci in the each HF are counted and shown in each panel (Magnification, × 100). (b) The increased number of cells with DNA damage foci in SG is rescued by BADGE. Epidermal sheets of p53^+/− mice and p53^TSD/− mice treated with or without BADGE were labeled for γ-H2A.X by IF and quantified by counting cells with DNA damage foci. n=30–40 HF for each genotype. Values represent means±S.D. and t-test was used to calculate the statistical significance, *P<0.05; NS, not significant. Arrowheads indicate cells with DNA damage foci in the SG while arrows indicate cells with DNA damage foci in HF under SG. (c) Immunoblotting for p16 was performed after extracting proteins from the skin of p53^+/− mice and p53^TSD/− mice treated with or without BADGE. (d) Epidermal sheets of p53^+/− and p53^TSD/− mice treated with or without BADGE were labeled by IF for p16 (red) as a senescent marker. (e) The relative expression level of PML in the skin of p53^+/−and p53^TSD mice treated with or without BADGE was determined by quantitative RT-PCR analysis. n=3. Values represent means±S.D. (f) The relative expression level of genes involved in adipogenesis in the skin of p53^TSD mice treated with or without BADGE were determined by quantitative RT-PCR analysis. n=3. Values represent means±S.D. (g) The increased number of cells with DNA damage foci in HF of p53^TSD/− mice are normalized by NAC treatment. Epidermal sheets of p53^+/− mice and p53^TSD/− mice treated with or without NAC were labeled for γ-H2A.X (red) by IF and representative images are shown. Cells with DNA damage foci in each HF are counted. n=30–40 HFs for each genotype. Values represent means±S.D. and t-test was used to calculate the statistical significance, *P<0.05 (Magnification, × 50). (h) Sebocyte progenitors depleted in p53^TSD/− mice are not rescued by ROS chelating. Epidermal sheets of p53^+/− mice or p53^TSD/− mice treated with or without NAC were labeled for Blimp1 (red) as a sebocyte progenitor marker by IF (Magnification, × 25)