Multiple outcomes of the germline p16INK4a mutation affecting senescence and immunity in human skin

Abstract

The integrated behaviour of multiple senescent cell types within a single human tissue leading to the development of malignancy is unclear. Patients with Familial Melanoma Syndrome (FMS) have heterozygous germline defects in the CDKN2A gene coding for the cyclin inhibitor p16^INK4a. Melanocytes within skin biopsies from FMS patients express significantly less p16^INK4a but express higher levels of the DNA-damage protein 𝛾H2AX a than fibroblastic cells. However, patient fibroblasts also exhibit defects since senescent cells do not increase in the skin during ageing and fibroblasts isolated from the skin of patients have increased replicative capacity compared to control fibroblasts in vitro, culminating in abnormal nuclear morphology. Patient derived fibroblasts also secreted less SASP than control cells. Predisposition of FMS patients to melanoma may therefore result from integrated dysregulation of senescence in multiple cell types in vivo. The inherently greater levels of DNA damage and the overdependence of melanocytes on p16 for cell cycle inhibition after DNA damage makes them exquisitely susceptible to malignant transformation. This may be accentuated by senescence-related defects in fibroblasts, in particular reduced SASP secretion that hinders recruitment of T cells in the steady state and thus reduces cutaneous immunosurveillance in vivo.

Keywords: SASP; cellular senescence; familial melanoma syndrome; immunology; p16; skin.

FIGURE 1

Decreased p16^INK4a expression by melanocytes in FMS patient skin. Formalin‐fixed‐paraffin‐embedded (FFPE) 5 mm skin biopsies from FMS patients (n = 10–11) and healthy, matched control donors (n = 11) were stained by immunofluorescence for p16^INK4a (green), Melan A (red) with DAPI nuclear counterstain (blue). Representative p16^INK4a staining shown in control and FMS patient skin (a) Yellow arrowheads indicate melanocytes. Graphs show (b) Average percentage of p16^INK4a+ keratinocytes; (c) Average percentage of p16^INK4a+ melanocytes; (d) Average percentage of p16^INK4a+ dermal fibroblastic cells; (e) Average epidermal cellular p16^INK4a MFI; (f) Average melanocyte p16^INK4a MFI; (g) Average dermal fibroblastic cell p16^INK4a MFI; (h) Correlation between p16^INK4a MFI and age in melanocytes; (i) Correlation between p16^INK4a MFI and age in dermal fibroblastic cells; (j) Correlation between melanocyte and dermal fibroblastic cell p16^INK4a MFI. Data are represented as mean ± SEM. Statistical significance calculated by unpaired two‐tailed Student's t‐test. ns, p > 0.05; *p ≤ 0.05.

FIGURE 2

Fewer TAF⁺ cutaneous melanocytes in FMS patients. Formalin‐fixed‐paraffin‐embedded (FFPE) 5 mm skin biopsies from FMS patients (n = 13–14) and healthy, matched control donors (n = 13–14) were cut into 3 μm sections and stained by immunoFISH for TAF and Melan A. (a) Representative high‐power confocal Z‐stack image of epidermis and superficial dermis showing telomeres (red intranuclear), Melan A (red membrane), 𝛾H2AX (green) with nuclear DAPI counterstain (blue). All Melan A+ cells from one skin section were imaged for each donor (20–30 cells per donor). Yellow arrowhead indicates colocalising TelC and 𝛾H2AX amounting to TAF. Inset white box shows a magnified TAF+ melanocyte, the dashed white line passing through a TAF is represented by histograms (b) showing the relative intensity of 𝛾H2AX (green line), Tel C (red line) across the distance of the white line. Graphs show (c) Mean number of Melan A⁺ cells/high power field; (d) FMS (n = 8), control donors (n = 9) mean numbers of nuclear 𝛾H2AX foci; (e) Percentage of cutaneous TAF⁺ cells; (f) Percentage of TAF⁺ melanocytes; (g) Percentage of TAF⁺ epidermal cells; (h) Percentage of TAF⁺ dermal fibroblastic cells; (i) Correlation between TAF⁺ melanocytes and age. Data are represented as mean ± SEM. ns, p > 0.05; *p ≤ 0.05; **p ≤ 0.01.

FIGURE 3

FMS fibroblasts upregulate p21 instead of p16^INK4a to become senescent, FMS melanocytes upregulate neither. (a–c) Cutaneous fibroblasts were isolated from healthy donor (n = 3–4) and FMS donor (n = 4) skin. They were cultured either in their non‐senescent state (NSEN) or induced to senescence with mitomycin C (MMC) or x‐ray irradiation (IR). 15 days after senescence‐induction, all cells were lysed and protein extracted for Western Blot. (a) Representative immunoblots. Graphs show mean grey value of detected protein normalized to GAPDH (B) p16^INK4a; (c) p21. (d–g) Cutaneous melanocytes were isolated from healthy donors (n = 3) and FMS donors (n = 3) skin. They were cultured either in their non‐senescent state (NSEN) or induced to senescence with mitomycin C (MMC). 15 days after senescence‐induction, all cells were fixed and stained for immunocytochemistry. Graphs show (d) Percentage of p16^INK4a+ melanocytes; (e) p16^INK4a MFI; (f) Percentage of p21⁺ melanocytes; (g) p21 MFI. Data are represented as mean ± SEM. Statistical significance calculated by Sidak's multiple comparisons test. *p ≤ 0.05; **p ≤ 0.01.

FIGURE 4

CDKN2A‐mutation attenuates fibroblast inflammatory protein secretion. Supernatants from healthy donor (n = 4–5) and FMS patients (n = 4) non‐senescent (NSEN) and senescent (IR) fibroblasts were cultured for 24 h and analysed for cytokine content using cytometric bead array. Graphs show concentrations of cytokines in supernatants as follows: (a) CXCL10; (b) CCL2; (c) CCL5; (d) IL‐8; (e) IL‐6; Data are represented as mean ± SEM. Statistical significance calculated by Dunn's multiple comparisons test, ns, p > 0.05; *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.0001.

FIGURE 5

Confirmation of replicative senescence in control and FMS patient fibroblasts. Cutaneous fibroblasts were isolated from healthy donor (n = 4–5) and FMS patients (n = 2–4) skin. They were serially passaged over an extended period until replicative senescence (RS) was achieved. (a) Representative images of SAβ‐Gal (blue) and (b) Ki67 staining (green). Graphs showing (c) Percentage of SAβ‐Gal+ cells; (d) Percentage of Ki67+ cells; Data are represented as mean ± SEM. Statistical significance calculated by Sidak's multiple comparisons test. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001.

FIGURE 6

Increased replicative capacity of FMS patient fibroblasts. Cutaneous fibroblasts were isolated from healthy donors (n = 5; ages 24, 30, 30, 65 and 66 years old) and FMS patients (n = 4; ages 26, 44, 45 and 68 years old) skin. They were serially passaged over an extended period until replicative senescence was achieved. (a) Graph showing cumulative population doublings over time; (b) Total population doublings per donor. (c) Pairwise comparisons of growth curves, the ages of the patients and controls are indicated in each panel and they were also gender matched. (d) Representative images showing morphological changes of senescent fibroblasts using DAPI nuclear stain (blue) and phalloidin actin stain (orange). Graphs showing (e) Percentage of 𝛾H2AX⁺ cells and (f) Percentage of Ki67⁺ cells in early‐passage and pre‐senescent fibroblasts. Data are represented as mean ± SEM. Statistical significance for (b) unpaired two‐tailed Student's t‐test, (c) permutation tests, (e, f) Sidak's multiple comparisons test. *p ≤ 0.05; **p ≤ 0.01; ****p ≤ 0.0001.

FIGURE 7

Reduced numbers of CD8⁺ and CD4⁺ T cells in FMS patient skin compared to healthy control skin. Formalin‐fixed‐paraffin‐embedded (FFPE) 5 mm skin biopsies from FMS patients and healthy, matched control donors were cut into 3 μm sections and stained by immunofluorescence for CD4 (white), CD8 (green), FoxP3 (red) with DAPI nuclear counterstain (blue). Representative staining shown in (a). We assessed the total number of lymphocytes (b) in FMS patients (n = 13) and controls (n = 10). We also assessed the number of CD8⁺ (c), CD4⁺ (d) and CD4⁺FoxP3⁺ T cells/mm² in FMS patients (n = 13) and healthy matched control donors (n = 22). Data are represented as mean ± SEM. Statistical significance calculated by unpaired two‐tailed Student's t‐test. ns, p > 0.05; **p ≤ 0.01.