Chronic ultraviolet B irradiation causes loss of hyaluronic acid from mouse dermis because of down-regulation of hyaluronic acid synthases

Abstract

Remodeling of the dermal extracellular matrix occurs during photoaging. Here, the effect of repetitive UVB irradiation on dermal hyaluronic acid (HA) was examined. C57/BL6 mice were chronically (182 days) irradiated with UVB, and consecutive skin biopsies were collected during the irradiation period and afterward (300 and 400 days of age). UVB caused marked loss of HA from the papillary dermis and down-regulation of HA synthase 1 (HAS1), HAS2, and HAS3 mRNA expression. In contrast, hyaluronidases (HYAL) 1, HYAL2, and HA receptor CD44 were unchanged. Furthermore, transforming growth factor beta-1 (TGF-beta1) and TGF-beta1-receptor II expression were decreased in UVB-irradiated biopsies, and TGF-beta1 strongly induced HAS1 and HAS2 expression in cultured dermal fibroblasts. Therefore, TGF-beta1 might be one factor involved in UVB-induced down-regulation of HAS enzymes. In addition, total cell number and the percentage of proliferating fibroblasts in the papillary dermis of UVB-irradiated mice were decreased. Down-regulation of HAS2 by lentiviral overexpression of short hairpin RNA in vitro caused inhibition of HA synthesis, DNA synthesis, and migration of dermal fibroblasts. In conclusion, chronic UVB irradiation induces loss of HA from the dermis, thereby contributing to the quiescent phenotype of dermal fibroblasts.

Figure 1

Progressive loss of HA from the dermis of chronically UVB-irradiated mice. Shaved C57/BL6 mice were either sham-irradiated or UVB-irradiated (three times per week, 3 MED ∼210 mJ/cm²), and consecutive skin biopsies were collected (see Materials and Methods). HA was detected by affinity histochemistry using biotinylated HAbP. A: Experimental design. B: HA staining. Top: Intrinsic aging; sham-irradiated controls at 150, 300, and 400 days. Bottom: Extrinsic aging; UVB-irradiated C57/BL6 mice at 150, 300, and 400 days. Arrows indicate the HA-rich zone in the papillary dermis where loss of HA is strongest. C: Densitometric quantification of HA staining, controls, n = 7, and UVB, n = 8; *P < 0.05 intrinsic aging versus UVB at 400 days (two-way analysis of variance). D: Biochemical quantification of HA in the dermis of controls versus UVB-irradiated skin at 400 days; mean ± SEM, controls, n = 4, and UVB, n = 5; *P < 0.05 (t-test). Original magnifications, ×100.

Figure 2

mRNA levels of HAS isoenzymes and BGN during intrinsic and UVB-induced aging. Shaved C57/BL6 mice were either sham-irradiated (control) or UVB-irradiated (three times per week, 3 MED ∼210 mJ/cm²), and consecutive skin biopsies were collected (see Materials and Methods). mRNA levels were estimated by real-time PCR and normalized to the expression in nonirradiated control mice at 150 days. HAS1 (A), HAS2 (B), HAS3 (C), and BGN (D) mRNA expression in nonirradiated and UVB-irradiated animals at the indicated times. Open bars represent nonirradiated controls, and black bars represent UVB-irradiated mice. Data are means ± SEM; controls, n = 8; UVB first biopsy, n = 8; UVB second biopsy, n = 7; UVB third biopsy, n = 6. P values were obtained by analysis of variance and Bonferroni’s post test to compare selected pairs; mean ± SEM; *P values <0.05.

Figure 3

mRNA levels of HYAL-1, HYAL-2, and CD44 during intrinsic and UVB-induced aging. Shaved C57/BL6 mice were either sham-irradiated (control) or UVB-irradiated (three times per week, 3 MED ∼210 mJ/cm²), and consecutive skin biopsies were collected (see Materials and Methods). mRNA levels were estimated by real-time PCR and normalized to the expression of nonirradiated control mice at 150 days. Hyal1 (A), Hyal-2 (B), and CD44 (C) mRNA expression in nonirradiated and UVB-irradiated animals at the indicated times. Open bars represent nonirradiated controls, and black bars represent UVB-irradiated mice. Data are means ± SEM, controls, n = 5, and UVB, n = 6. P values were obtained by analysis of variance and Bonferroni’s post test to compare selected pairs; mean ± SEM; P values >0.05.

Figure 4

mRNA levels of HAS isoenzymes in response to TGF-β1 in skin fibroblasts in vitro. Human skin fibroblasts were serum-starved for 24 hours and stimulated with either 1 or 10 ng/ml TGF-β1. mRNA levels of HAS isoforms were determined by real-time PCR at 6 and 24 hours. The data are normalized to the expression levels of serum-starved control fibroblasts. HAS1 (A and B), HAS2 (C and D), and HAS3 (E and F) mRNA at the indicated times after stimulation. Data are means ± SEM of n = 4 independent experiments in skin fibroblasts derived from two different donors. P values were calculated by analysis of variance and Bonferroni’s post test to compare selected pairs; mean ± SEM; *P values <0.05; **P values <0.01.

Figure 5

mRNA levels of TGF-β1 and TβR-II during intrinsic and UVB-induced aging. Shaved C57/BL6 mice were either sham-irradiated or UVB-irradiated (three times per week, 3 MED ∼210 mJ/cm²) and consecutive skin biopsies were collected (see Materials and Methods). mRNA levels were estimated by real-time PCR and normalized to the expression in nonirradiated control mice at 150 days. TGF-β1 (A) and TβR-II (B) mRNA expression in nonirradiated and UVB-irradiated animals at the indicated times. Open bars represent nonirradiated controls, and black bars represent UVB-irradiated mice. Data are means ± SEM; controls, n = 5, and UVB, n = 6. P values were obtained by two-way analysis of variance and Bonferroni’s post test; mean ± SEM; *P values <0.05.

Figure 6

Down-regulation of HAS2 by shRNA. Lentiviral transfection with shRNA targeting HAS2 (shHAS2) and empty control vector (pCL1) was performed in human skin fibroblasts. A: Five days after infection, analysis of HAS2 mRNA expression was performed using real-time RT-PCR. B: In parallel, HA secretion was determined by enzyme-linked immunosorbent assay, normalized to total protein and expressed as percentage of pCL1-transfected cells. C: Subsequently, on day 6 after infection, the migration toward 10% FCS was determined using a modified Boyden-chamber microchemotaxis assay on collagen type-1-coated membranes. D: [³H]Thymidine incorporation was determined on day 6 after transfection using 10% FCS as mitogenic stimulus. The data were normalized to total protein and represent the fold increase over starved controls. Data are means ± SEM of n = 4 independent experiments each performed in triplicates. P values were calculated by t-test; *P values <0.05.

Figure 7

Decreased proliferation of dermal skin fibroblasts after UVB irradiation. A and B: Double immunostaining of PCNA (red) and vimentin (green) of third biopsy samples; control (A) and UVB irradiation (B); arrowheads mark proliferating fibroblasts in A. Nuclei are stained by Hoechst 33324 in blue. C and D: Immunostaining of PCNA (red) and nuclear staining with Hoechst 33324 of third biopsy samples. To facilitate determination of the proliferative index, blue nuclear staining was turned into green to indicate PCNA-positive nuclei in the overlay in yellow; control (C) and UVB-irradiated (D); arrowheads mark proliferating cells. E: Proliferative index expressed as percent PCNA-positive nuclei. F: Total cell number in the papillary dermis determined as nuclei per μm². Data are means ± SEM; control, n = 9, UVB, n = 10. P values were calculated by t-test; *P value <0.05. Original magnifications: ×400 (A and B); ×200 (C and D).