- Email: [email protected]
RETRACTED: Paracrine cytokine interaction between UVB-exposed epidermal keratinocytes and dermal fibroblasts in stimulating expression of skin fibroblast-derived elastase
Cytokine 59 (2012) 166–175
Contents lists available at SciVerse ScienceDirect
Cytokine journal homepage: www.elsevier.com/locate/issn/10434666
Paracrine cytokine interaction between UVB-exposed epidermal keratinocytes and dermal fibroblasts in stimulating expression of skin fibroblast-derived elastase Hiroaki Nakajima, Ryosuke Yoshioka, Yoshiyuki Ezaki, Tomoyashu Nagai, Genji Imokawa ⇑ School of Bioscience and Biotechnology, Tokyo University of Technology, Japan
a r t i c l e
i n f o
Article history: Received 25 November 2011 Received in revised form 24 February 2012 Accepted 3 March 2012 Available online 14 April 2012 Keywords: Ultraviolet light Elastase Wrinkling Skin fibroblasts Neutral endopeptidase
a b s t r a c t Background: We recently reported that over-expression of skin fibroblast-derived elastase (SFE) plays a pivotal role in the mechanism of UVB-induced skin wrinkling. Since UVB penetrates only modestly to the dermis, we hypothesized that factors secreted by UVB-exposed keratinocytes in the epidermis trigger fibroblasts in the dermis to increase their expression of SFE which then degrades the elastic fibers. Objective: In this study, we characterized the paracrine interaction between human keratinocytes (HK) and human fibroblasts (HF) which leads to increased expression of SFE. Methods and results: Medium conditioned by UVB-exposed HK contained increased levels of IL-1a, GMCSF, IL-6, TNFa and IL-8. While HF cultured with those conditioned medium slightly down-regulated the gene expression of collagen and elastin, they significantly increased their expression of SFE at the transcriptional, translational and enzymatic levels. Neutralizing antibodies to IL-1a or GM-CSF significantly abolished the increased expression of SFE at the translational and/or enzymatic levels in HF cultured with those conditioned medium, while neutralizing antibodies to IL-6, IL-8 or TNFa had no such effect. The addition of IL-1a or GM-CSF, but not TNFa, IL-6 or IL-8, at concentrations ranging from 1 to 10 nm, significantly stimulated the enzymatic levels of SFE in HF. Conclusions: The sum of these findings suggests that IL-1a and GM-CSF are intrinsic cytokines secreted by UVB-exposed HK that stimulate expression of SFE by HF, leading to UVB-induced wrinkle formation. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Repetitive exposure to sunlight has been thought to be the most plausible factor that causes skin wrinkles since they frequently appear on the face. However, little is known about the precise physicochemical and biological mechanism(s) which lead to the formation of wrinkles due to sunlight exposure. In clinical studies, the formation of facial wrinkles has been associated with the loss of elastic properties of the skin [1,2]. Repetitive exposure of the skin to UVB radiation at suberythemal doses significantly reduces its elastic properties, which in turn leads to the formation of wrinkles [3]. Repetitive UVB irradiation of the skin also elicits a marked alteration in the microstructure of collagen fibers [4–6] as well as the three-dimensional structure of elastic fibers, which are closely linked to a subsequent reduction in elastic properties [7,8]. The Abbreviations: GM-CSF, granulocyte macrophage colony stimulatory factor; IL1a, Interleukin-1a; NEP, neutral endopeptidase; SFE, skin fibroblast-derived elastase; STANA, N-succinyl-tri-alanyl-p-nitroaniline; UVB, ultraviolet B; MMP-1, matrix-meltalloproteinase-1. ⇑ Corresponding author. Address: School of Bioscience and Biotechnology, Tokyo University of Technology, 1404-1 Katakura, Hachioji, Tokyo 192-0982, Japan. Tel./fax: +81 42 637 2424. E-mail address: [email protected] (G. Imokawa). 1043-4666/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.cyto.2012.03.008
alteration of collagen bundles and elastic fiber architecture has been substantiated by a marked continuous up-regulation of a collagen- or an elastin-degrading enzymes, collagenase-I (matrixmetalloproteinase (MMP)-I or elastase in wrinkled skin after repetitive UVB irradiation [9–12]as well as after ovariectomy [13]. A study to determine which matrix protease(s) are linked to the loss of skin elasticity in the dermis of UVB-irradiated mouse skin reported a significant increase in the activity of elastase in UVBirradiated skin compared with non-irradiated skin [13]. In contrast, there was a slight but not significant increase in the activity of collagenase I in the dermis at week 14 in UVB-irradiated skin compared with non-irradiated skin [13]. On the other hand, there was a decrease in the activity of collagenase IV in UVB-irradiated skin at the same time point compared with non-irradiated skin [13]. On the other hand, UVA-irradiation elicited skin sagging like age-related morphological structure of the human cheek [14] rather than wrinkling as a result of higher up-regulation of MMP-1 expression compared with skin elastase expression in the dermis [13] and in cultured human fibroblasts [15]. Thus, we hypothesized that enhanced elastase activity in the dermis of UVB-irradiated skin plays an important role in the degeneration of elastic fibers, which eventually results in the loss of skin elasticity. This elastase-triggered alteration of the three-dimensional
H. Nakajima et al. / Cytokine 59 (2012) 166–175
structure of elastic fibers was validated by our UVB studies [16,17] which used a synthetic inhibitor specific for skin fibroblast elastases to demonstrate a close interrelationship among wrinkle formation, elastic properties and elastic fiber linearity, a marker for the alteration of elastic fibers. The fact that elastase inhibitors can serve as anti-wrinkling agents was also substantiated by another animal study [18] and by 1 year clinical study of human volunteers [19]. The sum of this evidence strongly suggests that enhanced elastase activity produced by dermal fibroblasts plays a pivotal role in the mechanism of UVB-induced wrinkling of the skin. Skin fibroblast-derived elastase (SFE) is a 94 kDa membranebound type metalloprotease with a neutral optimum pH [20–23]. Although there were several candidate enzymes for SFE, such as 92, 72 kDa type IV collagenase (gelatinase) [24,25], neutrophil elastase [26], cathepsin G [27] and protease 3 [28], none of those had properties which matched the characteristics of SFE. Since there are similarities between SFE and neutral endopeptidase (NEP) [29] in terms of their size (MW 97,000), their membranebound nature, and their inhibitory profiles, we used immunoprecipitation and Western blotting with an antibody to NEP to show that SFE is identical to NEP (here after referred as to SFE) [30]. The identification of SFE as NEP enabled us to clarify epithelialmesenchymal interactions, especially at the gene and protein levels, whereby UVB irradiation that penetrates only the epidermis [31] causes fibroblasts in the dermis to stimulate their expression of NEP. Thus, we hypothesized that soluble cytokines secreted by UVB-exposed keratinocytes pass through the basement membrane into the dermis and cause dermal fibroblasts to increase their expression of NEP which then disrupts the three dimensional networks of elastic fibers. This results in the loss of skin elastic properties, an essential parameter that leads to wrinkle formation. In this study, we characterized the paracrine linkage between UVB-exposed epidermal keratinocytes and dermal fibroblasts that leads to wrinkle formation in the skin. We determined factors involved in the up-regulation of fibroblast elastases by examining the stimulatory effects of conditioned medium from UVB-exposed human keratinocytes on fibroblast elastases and the inhibitory effects of neutralizing antibodies to various paracrine factors. We report that interleukin-1a and granulocyte macrophage colony stimulatory factor are intrinsic cytokines secreted by UVB-exposed human keratinocytes that stimulate the expression of NEP by human fibroblasts.
167
2. Materials and methods 2.1. Materials Human primary keratinocytes (HPK) (derived from adult abdomen skin, age 32, female, Caucasian), normal human fibroblasts (HF) (derived from foreskin, Caucasian), serum-free keratinocyte growth medium (Medium 154S) containing low calcium (0.2 mM), bovine pituitary extract (BPE), and epidermal growth factor (EGF), were obtained from Kurabo (Tokyo, Japan). Human keratinocyte cell line, HaCaT cells were kindly supplied by Dr. M. Furue (Kyushu University, School of Medicine, Department of Dermatology). Anti-human NEP (CD10:SN5C: sc-19586) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Neutralizing antibodies such as anti-human IL-6 monoclonal mouse antibody (MAB206), anti-human CXCL8/IL-8 monoclonal mouse antibody (MAB208), anti-human TNF-a monoclonal mouse antibody (MAB2101) and anti-human GM-CSF monoclonal mouse antibody (MAB215) were purchased from R&D Systems (Minneapolis, MN, USA). The neutralizing mouse antibody to human IL-1a (hIL-1a)(L122M) and normal mouse IgG (12-371) was obtained from EXALPHA Biologicals (Shirley, MA, USA) and Millipore Co. Ltd., respectively. Recombinant human IL-6 (206-IL-010), recombinant human CXCL8/IL-8 (618-IL-010) and recombinant human GM-CSF (215-GM-010) were purchased from R&D Systems. Recombinant human IL-1a (200-01A) and recombinant human TNFa (300-01A) were obtained from PEPROTECH (London, UK). ELISA Kits (for IL-1a, IL-6, IL-8, GM-CSF and TNFa) were obtained from Endogen (Thermo Fisher Scientific, Yokohama, Japan). Horseradish peroxidase-conjugated goat polyclonal anti-mouse IgG was obtained from Transduction Laboratories (Franklin Lakes, NJ, USA). Other chemicals of reagent grade were purchased from Sigma-Aldrich (St. Louis, MO, USA). 2.2. Cell cultures HF were cultivated as previously described [15]. HPK were maintained in serum-free keratinocyte medium (Medium 154S) (Kurabo, Tokyo Japan) supplemented with 5 ng/ml EGF and 50 lg/ml BPE at 37 °C with 5% CO2. HaCaT cells were maintained in Dulbecco modified Eagle’s medium (DMEM) with 10% FCS, 100 lg/ml penicillin, 100 lg/ml streptomycin, and 250 ng/ml
Fig. 1. Effects of conditioned medium from UVB-exposed HPK on mRNA levels of NEP, elastin, collagenase I and collagen in HF as revealed by real-time RT-PCR analysis. The conditioned medium from UVB-exposed HPK was added at 1:1 dilution to cultured HF to measure expression levels of each mRNA at 4 h post-incubation. Values are means ± S.D. derived from three independent experiments. ⁄p < 0.05. ⁄⁄p < 0.01.
168
H. Nakajima et al. / Cytokine 59 (2012) 166–175
Fig. 2. Effects of conditioned medium from UVB-exposed HPK or HaCaT cells on NEP protein (A) and enzymatic levels (B) in HF. (A): NEP protein levels were measured by Western blotting analysis 24, 48 and 72 h after the addition of conditioned medium (CM) from UVB (0 or 80 mJ/cm2)-exposed HaCaT cells or non-conditioned fresh DMEM medium (NCM) at 1:1 dilution to cultured HF. Representative immunoblots from three independent experiments are shown. (B): (a) HPK; (b) HaCaT cells. NEP activity levels were measured by enzymatic assay using STANA 72 h after the addition of the condition medium at 1:1 dilution to cultured HF. Values are means ± S.D. derived from three independent experiments. ⁄p < 0.05. ⁄⁄p < 0.01.
amphotericin B at 37 °C in a 95% air, 5% CO2 atmosphere. In experiments to measure cytokine levels by ELISA kits and to collect conditioned medium after UVB irradiation, HPK or HaCaT cells were cultured in 10 cm dishes with 5 ml of Medium 154S without EGF and BPE or with 5 ml of DMEM without FCS, respectively. For experiments to test paracrine interactions, the keratinocyte-conditioned medium was obtained by culturing HPK in 10 cm dishes with 5 ml of Medium 154S without EGF and BPE or HaCaT cells in 10 cm dishes with 5 ml of DMEM without FCS for 3 days after UVB irradiation. HF were seeded in 6 cm dishes at a density of approximately 5 105 cells with DMEM with 10% FCS and cultured
for 24 h. After 24 h starvation in DMEM without FCS, 1.0 ml of the medium conditioned by UVB exposed keratinocytes was added at 1:1 dilution to 1.0 ml of fresh culture medium (DMEM) of HF. 2.3. UVB irradiation HPK or HaCaT cells cultured at approximately 1.5–2.0 106 cells in 10 cm dishes were washed twice with PBS and were then exposed to UVB. The UVB source was a SE fluorescent lamp (Clinical Supply, Tokyo, Japan) that emitted an energy spectrum with high fluency in the UVB region (280–320 nm) with a peak at
H. Nakajima et al. / Cytokine 59 (2012) 166–175
169
Fig. 3. Secretion levels of cytokines (pg/ml) in the medium from HaCaT cells or HPK 72 h after UVB irradiation at 0–80 mJ/cm2. (a) HaCaT cells; (b) HPK. Cultured HaCaT cells or HPK were exposed to UVB at the indicated doses and secretion levels of cytokines were measured by ELISA kit at 72 post-irradiation. Values are means ± S.D. derived from three independent experiments. ⁄⁄p < 0.01. ⁄p < 0.05 (vs. 0 mJ/cm2).
305 nm. The emitted dose was calculated using a UVB radiometer photodetector (Torex, Tokyo, Japan). After UVB irradiation, the cells were maintained for 48–72 h in 5 ml of DMEM without FCS (for HaCaT cells) or in 5 ml of Medium 154S without EGF and BPE (for HPK) at 37 °C in a 95% air, 5% CO2 atmosphere. Preliminary testing determined no significant effect on cell viability of HPK and HaCaT cells by UVB irradiation with 80 mJ/cm2 at 48 and 72 h post-irradiation as measured by MTT assay [32]. 2.4. Neutralizing antibody experiments After 1.0 ml of the conditioned medium obtained from UVB- or mock-exposed HaCaT cells or HPK were added to 1.0 ml of fresh DMEM, neutralizing antibodies diluted in PBS were added at the indicated concentrations into the diluted conditioned medium. After replaced with the prepared conditioned medium, HF were maintained to culture for 72 h in 6 cm dishes. 2.5. Measurement of elastase activity Fibroblast enzyme solution was prepared as previously described [15]. Elastase activity using the synthetic substrate N-succinyl-tri-alanyl-p-nitroaniline (STANA) (Peptide Institute Inc., Osaka, Japan) was measured as previously described [33]. The release of p-nitroaniline was measured by absorbance at 405 nm and enzymatic activity is expressed as nmol p-nitroaniline/h/lg protein. Finally a comparison was performed by expressing activity as relative to elastase activity.
to X-ray film for the specified times to detect bands, and specific bands were quantitated by densitometric scanning and were analyzed with Quantity OneÒ (Bio Rad). 2.7. Real-time RT-PCR analysis Levels of NEP, MMP-1, collagen and elastin mRNAs in HF were examined using real-time RT-PCR as previously described [15]. Primers with the following sequences were used: GAPDH: Forward, 50 -GAAGGTGAAGGTCGGAGTCAACG-30 , Reverse, 50 -AGTCCTTCCACGATAACCAAAGTTG-30 384 bp, NEP: Forward, 50 -GAACCTACAAGGAGTCCAGA-30 Reverse, 50 -GCTCCACTTATCCACTCATC-30 121 bp, Collagenase I (MMP-1): Forward, 50 -GCTGGGAGCAAACACATCTGAGGT-30 , Reverse, 50 -TGAGCCGCAACACGATGTAAGTTG-30 331 bp, Elastin: Forward, 50 -AGGTGTATACCCAGGTGGCGTGCT-30 , Reverse, 50 -CAACCCCTGTCCCTGTTGGGTAAC-30 311 bp, Collagen: Forward, 50 -CTGGTCCCCAAGGCTTCCAAGGTC-30 , Reverse, 50 -CCATCATTTCCACGAGCACCAGCA-30 391 bp. Preliminary testing determined no significant effect on GAPDH levels in HPK and HaCaT cells by UVB irradiation with 80 mJ/cm2 at 4–24 h post-irradiation. 2.8. Statistical analysis All data are expressed as means ± SD (n = 3). For pairwise comparisons, either Student’s t-test or Welch’s t-test was applied. For multiple comparisons, data were tested by one-way ANOVA, and subsequently using the Tukey or Dunnet multiple comparison test. P values less than 0.05 are considered statistically significant. 3. Results
2.6. Western blotting Cell lysates were separated on 10% SDS-PAGE gels, and were blotted to PVDF membranes (Bio-Rad, Hercules, CA, USA). After blocking with 3% bovine serum albumin-containing Tris/HCl (pH 7.5, 100 mM NaCl), the membranes were treated with the mouse monoclonal anti-human NEP antibody at room temperature for 1 h, then with anti-mouse IgG conjugated to horseradish peroxidase, and finally with Amersham ECL reagents. Plots were exposed
3.1. Effects of conditioned medium from UVB-exposed human primary keratinocytes on mRNA levels in human fibroblasts To test the paracrine cytokine mechanism underlying UVB-induced wrinkle formation, we asked initially whether conditioned medium from UVB-exposed human primary keratinocytes (HPK) would stimulate the expression of genes encoding matrix proteins or matrix metalloproteases in human fibroblasts (HF). Real-time
170
H. Nakajima et al. / Cytokine 59 (2012) 166–175
Fig. 4. Neutralizing antibody abrogates the stimulated NEP activity. (A): HaCaT cells, (a) anti-human IL-1a antibody, (b) anti-human GM-CSF antibody, (c) anti-human IL-6 antibody, (d) anti-human IL-8 antibody, (e) anti-human TNFa antibody, (f)non-specific mouse IgG. (B): HPK, (a) anti-human IL-1a antibody, (b) anti-human GM-CSF antibody. Antibodies as well as non-specific mouse IgG were incubated at the indicated concentrations with the two fold diluted conditioned medium. After replaced with the prepared conditioned medium, HF were maintained to culture for 72 h. Lysates of HF were then assessed for NEP activity using STANA. Values are means ± S.D. derived from three independent experiments. ⁄⁄p < 0.01. ⁄p < 0.05.
RT-PCR for NEP mRNA revealed that when the HPK-conditioned medium was added at 1:1 dilution to cultured HF, levels of NEP mRNA were significantly stimulated at 4 post-incubation with a UVB dose of 80 mJ/cm2 compared with mock-irradiated controls (Fig. 1). Real-time RT-PCR for elastin mRNA revealed that the conditioned medium from UVB-exposed HPK slightly down-regulated elastin mRNA levels at 4 h post-incubation with UVB doses of 60 or 80 mJ/cm2 (Fig. 1). Real-time RT-PCR for collagenase type I mRNA revealed that the conditioned medium from UVB-irradiated HPK significantly stimulated collagenase type I (MMP-1) mRNA levels at 4 h post-incubation with a UVB dose of 80 mJ/cm2 (Fig. 1). Real-time RT-PCR for collagen mRNA revealed that the conditioned medium from UVB-irradiated HPKslightly down-regulated collagen mRNA levels at 4 h post-incubation with UVB doses of 60 or 80 mJ/cm2 (Fig. 1). 3.2. Effects of conditioned medium from UVB-exposed HaCaT cells on NEP protein levels in HF Since mRNA expression levels for NEP in HF were distinctly stimulated by the conditioned medium from UVB-exposed human keratinocytes, we next asked whether the conditioned medium stimulates NEP protein levels in HF. Western blotting analysis for
NEP protein revealed that when the conditioned medium from UVB-exposed HaCaT cells at a dose of 80 mJ/cm2 was added at 1:1 dilution to cultured HF, NEP protein levels were markedly stimulated at 72 h post-incubation as compared with the mockirradiation controls (Fig. 2A). 3.3. Effects of conditioned medium from UVB-exposed human keratinocytes on NEP activity in human fibroblasts Since the NEP mRNA and protein levels in HF were significantly stimulated by the conditioned medium from UVB-exposed human keratinocytes, we next asked whether the conditioned medium stimulates the enzymatic activity of NEP in HF. Enzymatic assays for NEP revealed that when the conditioned medium obtained from UVB-exposed HaCaT cells and from UVB-exposed HPK was added at 1:1 dilution to cultured HF, the enzymatic activity of NEP were significantly stimulated at 72 h post-incubation as compared with the mock-irradiation controls (Fig. 2B). 3.4. Cytokine secretion in UVB-exposed human keratinocytes When cultured HPK or HaCaT cells were exposed to UVB at the doses of 0–80 mJ/cm2 and cytokines secreted into the conditioned
H. Nakajima et al. / Cytokine 59 (2012) 166–175
171
Fig. 4 (continued)
medium were measured by ELISA kits, IL-1a, IL-6, IL-8, and GMCSF were significantly increased at 72 h post-irradiation (Fig. 3).
3.7. Effects of cytokines on the enzymatic activity of NEP in human fibroblasts
3.5. Neutralizing antibody abrogates the stimulated NEP activity
We next examined the potential of secreted cytokines to stimulate the enzymatic activity of NEP in HF. When HF were treated with IL-1a, GM-CSF, IL-6 or IL-8 at 0, 1, 5, and 10 nm, the enzymatic activity of NEP was significantly stimulated over 72 h by incubation with IL-1a or GM-CSF, but not by IL-6, IL-8 or TNFa (Fig. 6).
Since HF cultured with the conditioned medium from UVB-exposed HPK significantly increased the expression of NEP at the transcriptional, translational and enzymatic levels, we have attempted to identify kinds of cytokines mainly responsible for the stimulated expression of NEP by neutralizing experiments with antibodies to cytokines secreted in the conditioned medium. When conditioned medium obtained from UVB- or non-exposed HaCaT cells was treated with neutralizing antibodies to IL-1a, GM-CSF, IL-6, TNFa or IL-8 or with a non-specific mouse IgG, the stimulation of NEP activity was markedly abrogated by anti-IL-1a and anti-GM-CSF antibodies but not by anti-IL-6, anti-IL-8 or anti-TNFa or the non-specific mouse IgG (Fig. 4A). When conditioned medium obtained from UVB- or non-exposed HPK was treated with anti-IL-1a or anti GM-CSF antibodies, the stimulation of NEP activity was markedly abrogated for UVB-exposed HPK but not for non-exposed cells by anti-IL-1a and by anti-GM-CSF antibodies (Fig. 4B). 3.6. Neutralizing antibody abrogates the stimulated NEP protein When conditioned medium obtained from UVB- or non-exposed HaCaT cells was treated with IL-1a, GM-CSF, IL-6, or IL-8 antibodies or with non-specific mouse IgG, the increased level of NEP protein was significantly abrogated by IL-1a and GM-CSF antibodies, but not by IL-6, IL-8, TNFa antibodies or by a non-specific mouse IgG (Fig. 5).
4. Discussion Although the enhanced activity of NEP in skin fibroblasts is responsible for the UVB-induced cascade of biological events that leads to the degeneration of elastic fibers and the reduced skin elasticity, which in turn leads to wrinkle formation [17], little is known about the mechanism(s) whereby NEP activity in dermal fibroblasts is stimulated by UVB irradiation. Based on the fact that UVB penetrates to the dermis to only a minor extent [31] and that UVB down-regulates NEP expression in HF (unpublished data), the most plausible mechanism is that UVB elicits the production of a factor(s) in the epidermis which then stimulates the expression of elastases by fibroblasts in the dermis. To test that hypothesis, we asked whether conditioned medium from UVB-exposed human keratinocytes stimulates the expression of matrix proteins or matrix metalloproteases, especially NEP, in HF. Real-time RT-PCR analysis revealed that the conditioned medium from UVB-exposed human keratinocytes stimulates NEP mRNA levels and collagenase type I mRNA levels. In contrast, levels of mRNAs encoding elastin as well as collagen are down-regulated at 4 h post-incubation.
172
H. Nakajima et al. / Cytokine 59 (2012) 166–175
Fig. 5. Neutralizing antibody abrogates the stimulated NEP protein. (A): anti-human IL-1a antibody, (B): anti-human GM-CSF antibody, (C): anti-human IL-6 antibody, (D): anti-human IL-8 antibody, (E): anti-human TNFa antibody, (F): non-specific mouse IgG. Antibodies as well as non-specific mouse IgG were added at the indicated concentrations to the two fold diluted conditioned medium. After replaced with the prepared conditioned medium, HF were maintained to culture for 72 h. Lysates of HF were then subjected to Western blotting analysis. Values are means ± S.D. derived from three independent experiments. ⁄⁄p < 0.01. ⁄p < 0.05.
Western blotting and enzymatic assay of HF treated with conditioned medium from UVB-exposed human keratinocytes demonstrated that NEP protein and enzymatic levels in human fibroblasts are significantly increased at 48 or 72 h post-incubation. The sum of these findings indicates that paracrine factors are secreted by UVB-exposed human keratinocytes which penetrate into the dermis to trigger dermal fibroblasts to stimulate NEP activity. Comparison of the in vivo profiles for matrix
proteins/metallo-proteinases in UVB-exposed and wrinkled skin and the in vitro cellular effects of conditioned medium from UVB-exposed human keratinocytes on gene expression patterns in HF (in both of which, while there is a down-regulated expression of elastin mRNA, NEP and MMP-1mRNAs are markedly up-regulated) [17] revealed that the effects of the conditioned medium on HF mimics the in vivo situation in wrinkled skin despite the fact that the frequency of UVB exposure differs greatly between the
H. Nakajima et al. / Cytokine 59 (2012) 166–175
173
Fig. 6. Effects of cytokines on the enzymatic activity of SFE HF as revealed by enzymatic assay using STANA. NEP activity levels were measured by enzymatic assay using STANA 72 h after the addition of cytokines to cultured HF at the indicated concentrations. Values are means ± S.D. derived from three independent experiments. IL-1a (a), GM-CSF (b), IL-6 (c), IL-8 (d), TNFa (e). ⁄⁄p < 0.01.
Fig. 7. A scheme for wrinkling mechanism. Repetitive UVB exposure causes keratinocytes to secrete IL-1a which triggers GM-CSF secretion in an autocrine fashion. The secreted IL-1a and GM-CSF penetrate the dermis to stimulate the expression of NEP by fibroblasts, which then cleaves elastic fibers surrounding the fibroblasts, leading to deterioration of the three-dimensional configuration of elastic fibers. This results in a loss of skin elasticity and in turn leads to wrinkle formation.
174
H. Nakajima et al. / Cytokine 59 (2012) 166–175
in vitro and in vivo situations. This suggested that the enhanced activity of NEP or MMP-1 in UVB-irradiated skin [12,13] is mediated by basement membrane-permeable soluble factors secreted by UVB-exposed human keratinocytes. Taken together, the sum of available evidence supports our hypothesis for a mechanism of wrinkle formation by which cytokines are released by keratinocytes following UVB irradiation, triggering dermal fibroblasts to stimulate the expression and activity of NEP. The enhanced NEP activity then results in the deterioration of the three-dimensional architecture of elastic fibers, reducing skin elasticity, and eventually leading to the formation of wrinkles. To determine which cytokines secreted by UVB-exposed human keratinocytes are intrinsic soluble factors that cause HF to stimulate the activity of NEP, we used conditioned medium to evaluate the inhibitory effects of several neutralizing antibodies to cytokines or a chemokine secreted by UVB-exposed human keratinocytes on NEP expression in HF. Those results revealed that among the neutralizing antibodies tested, the anti-IL-1a and anti-GM-CSF (but not anti-IL-6, anti-IL-8 or anti-TNFa or the non-specific mouse IgG) significantly abolished the enhanced activity of NEP in HF after the addition of the conditioned medium of UVB-exposed human keratinocytes. The neutralizing effects of the anti-IL-1a and the anti-GM-CSF antibodies on enzymatic activity levels were accompanied by the down-regulated protein expression of NEP. The sum of these findings strongly suggests that IL-1a and GM-CSF are intrinsic keratinocyte-derived cytokines associated with the up-regulated expression of NEP. Interestingly, those two antibodies also reduced the activity of NEP after the addition of the conditioned medium from non-exposed HaCaT cells. These findings indicate that the conditioned medium from non-exposed HaCaT cells also contains IL-1a and GM-CSF at concentrations sufficient, although to a lesser extent, to stimulate NEP activity in HF. Before speculating about the paracrine cytokine mechanism involved in the activation of NEP in vivo, it should be noted that studies using conditioned medium have several circumstances distinct from the in vivo situation; (1) The conditioned medium used human keratinocytes exposed to single UVB in contrast to the repeated UVB exposure under in vivo conditions required for wrinkle formation [11,12]. (2) The conditioned medium contains accumulated levels of cytokines that were secreted for 72 h postirradiation, which differs from the in vivo situation where shortly after UVB irradiation, secreted cytokines soon penetrate the epidermis to stimulate fibroblasts in the dermis, leading to the enhanced expression of NEP. (3) Since several cytokines secreted by UVB-exposed keratinocytes result from autocrine loop stimulation by primarily released cytokines such as IL-1a [34], the concentration of each cytokine in the conditioned medium may differ, which may not precisely reflect the repeated and accumulated secretion of NEP-inducing cytokines under the in vivo long term conditions required for wrinkle formation. In spite of these reservations, our data clearly show that IL-1a and GM-CSF are the only cytokines whose antibodies have specific neutralizing effects on the stimulated activity or protein level of NEP in HF cultured with the conditioned medium from UVB-exposed human keratinocytes. Consistent with this, they are found to be the cytokines capable of significantly stimulating the enzymatic activity in HF although it remains to be clarified whether IL-1a serves in an autocrine loop to trigger the secretion of GM-CSF and/or could directly cause skin fibroblasts to stimulate the activity of NEP.
5. Conclusion Thus, as depicted in Fig. 7, we propose a UVB-induced wrinkling mechanism as follows: Repetitive UVB exposure causes keratino-
cytes to secrete IL-1a which triggers GM-CSF secretion in an autocrine fashion. The secreted IL-1a and GM-CSF penetrate the dermis to stimulate the expression of NEP by fibroblasts, which then cleaves elastic fibers surrounding the fibroblasts [23], leading to deterioration of the three-dimensional configuration of elastic fibers. This results in a loss of skin elasticity and in turn leads to wrinkle formation. Conflict of interest The authors have declared that no conflict of interest exists. References [1] Akasaki S, Imokawa G. Mechanical methods for evaluating skin surface architecture in relation to wrinkling. J Dermatol Sci 2001;27(Suppl. 1):S5–S10. [2] Akazaki S, Nakagawa H, Kazama H, Osanai O, Kawai M, Takema Y, et al. Agerelated changes in skin wrinkles assessed by a novel three-dimensional morphometric analysis. Br J Dermatol 2002;147:689–95. [3] Takema Y, Imokawa G. The effects of UV-A and UV-B irradiation on the viscoelastic properties of hairless mouse skin in vivo. Dermatology 1998;196:397–400. [4] Schwartz E. Connective tissue alterations in the skin of ultraviolet irradiated hairless mice. J Invest Dermatol 1988;91:158–61. [5] Zheng P, Kligman LH. UVA-induced ultrastructural changes in hairless mouse skin: a comparison to UVB-induced damage. J Invest Dermatol 1993;100:194–9. [6] Rabe JH, Mamelak AJ, McElgunn PJS, Morison WL, Sauder DN. Photoaging: mechanisms and repair. J Am Acad Dermatol 2006;55:1–19. [7] Imayama S, Nakamura K, Takeuchi M, Hori Y, Takema Y, Sakaino Y, et al. Ultraviolet-B irradiation deforms the configuration of elastic fibers during the induction of actinic elastosis in rats. J Dermatol Sci 1994;7:32–8. [8] Imokawa G, Takema Y, Yorimoto Y, Tsukahara K, Kawai M, Imayama S. Degree of ultraviolet-induced tortuosity of elastic fibers in rat skin is age-dependent. J Invest Dermatol 1995;105:254–8. [9] Fisher GJ, Datta SC, Talwar HS, Wang ZQ, Varani J, Kang S, et al. Molecular basis of sun-induced premature skin ageing and retinoid antagonism. Nature 1996;379:335–9. [10] Brennan M, Bhatti H, Nerusu KC, Bhagavathula N, Kang S, Fisher GJ, et al. Matrix metalloproteinase-1 is the major collagenolytic enzyme responsible for collagen damage in UV-irradiated human skin. Photochem Photobiol 2003;78:43–8. [11] Tsukahara K, Takema Y, Moriwaki S, Tsuji N, Suzuki Y, Fujimura T, et al. Selective inhibition of skin fibroblast elastase elicits a dose-dependent prevention of UVB-induced wrinkle formation. J Invest Dermatol 2001;117:671–7. [12] Tsuji N, Moriwaki S, Suzuki Y, Takema Y, Imokawa G. The role of elastases secreted by fibroblasts in wrinkle formation: implication through selective inhibition of elastase activity. Photochem Photobiol 2001;74:283–90. [13] Tsukahara K, Nakagawa H, Moriwaki S, Shingo K, Ohuchi A, Takema Y, et al. Ovariectomy is sufficient to accelerate spontaneous skin ageing and to stimulate ultraviolet irradiation-induced photoageing of murine skin. Br J Dermatol 2004;151:984–94. [14] Tsukahara K, Takema Y, Fujimura T, Moriwaki S, Kitahara T, Imokawa G. Determination of age-related changes in the morphological structure (sagging) of the human cheek using a photonumeric scale and three-dimensional surface parameters. Int J Cosmet Sci 2000;22:247–58. [15] Suganuma K, Nakajima H, Otsuki M, Imokawa G. Astaxanthin attenuates the UVA-induced up-regulation of matrix-metalloproteinase-1 and skin fibroblast elastase in human dermal fibroblasts. J Dermatol Sci 2010;58:136–42. [16] Imokawa G. Recent advances in characterizing biological mechanisms underlying UV-induced wrinkles: a pivotal role of fibrobrast-derived elastase. Arch Dermatol Res 2008;300(Suppl. 1):S7–S20. [17] Imokawa G. Mechanism of UVB-induced wrinkling of the skin: paracrine cytokine linkage between keratinocytes and fibroblasts leading to the stimulation of elastase. J Invest Dermatol Symp Proc 2009;14:36. [18] Tsukahara K, Nakagawa H, Moriwaki S, Takema Y, Fujimura T, Imokawa G. Inhibition of ultraviolet-B-induced wrinkle formation by an elastase-inhibiting herbal extract: implication for the mechanism underlying elastase-associated wrinkles. Int J Dermatol 2006;45:460–8. [19] Tsukahara K, Nakagawa H, Ohtuki M, Moriwaki S, Fujimura T, Takema Y, et al. A clinical effect of Zingiber officinale (L.) rose on facial wrinkles and skin elasticity. J Mol Biol 2004;27:19–27. [20] Godeau G, Hornebeck W. Morphometric analysis of the degradation of human skin elastic fibers by human leukocyte elastase (EC 3–4-21-37) and human skin fibroblast elastase (EC 3–4-24). Pathol Biol 1988;36:1133–8. [21] Homsy R, Pelletier-Lebon P, Tixier JM, Godeau G, Robert L, Hornebeck W. Characterization of human skin fibroblast elastase activity. J Invest Dermatol 1988;91:472–7. [22] Croute F, Delaporte E, Bonnefoy JY, Fertin C, Thivolet J. Nicolas JF Interleukin-1 beta stimulates fibroblast elastase activity. Br J Dermatol 1991;124:538–41.
H. Nakajima et al. / Cytokine 59 (2012) 166–175 [23] Mecham RP, Broekelmann TJ, Fliszar CJ, Shapiro SD, Welgus HG, Senior RM. Elastin degradation by matrix metalloproteinases. J Biol Chem 1997;272:18071–6. [24] Senior RM, Griffin GL, Fliszar CJ, Shapiro SD, Goldberg GI, Wegus HG. Human 92- and 72-kilodalton type IV collagenase are elastase. J Biol Chem 1991;266:7870–5. [25] Shipley JM, Doyle GAR, Fliszar CJ, Ye Q-Z, Johnson LL, Shapiro SD, et al. The structural basis for the elastolytic activity of the 92-kDa and 72-kDa gelatinases: role of the fibronectin type II-like repeats. J Biol Chem 1996;271:4335–41. [26] Cambell EJ, Cambell MA. Pericellular proteolysis by neutrophils in the presence of proteinase inhibitors: effects of substrate opsonization. J Cell Biol 1988;106:667. [27] Boudier C, Godeau G, Hornebeck W, Robert L, Bieth JG. The elastolytic activity of cathepsin G: an ex vivo study with dermal elastin. Am J Respir Cell Mol Biol 1991;4:497. [28] Kao RC, Wehner NG, Skubitz KM, Gray BH, Hoidal JR. Proteinase 3. A distinct human polymorphonuclear leukocyte proteinase that produces emphysema in hamsters. J Clin Invest 1988;82:1963–73.
175
[29] Fulcher IS, Matsas R, Turner AJ, Kenny AJ. Kidney neutral endopeptidase and the hydrolysis of enkephalin by synaptic membranes show similar sensitivity to inhibitors. Biochem J 1982;203:519–22. [30] Morisaki N, Moriwaki S, Sugiyama-Nakagiri Y, Haketa K, Takema Y, Imokawa G. Neprilysin is identical to skin fibroblast elastase – its role in skin ageing and UV responses. J Biol Chem 2010;285:39819–27. [31] Bruls WA, Slaper H, van der Leun JC, Berrens L. Transmission of human epidermis and stratum corneum as a function of thickness in the ultraviolet and visible wavelengths. Photochem Photobiol 1984;40:485–94. [32] Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63. [33] Nakagawa K, Tsuji T, Kadoya A, Hamada T. Elastase-like enzyme activity in cultured human fibroblast. Skin Re 1987;29:793–7. [34] Imokawa G, Yada Y, Miyagishi M. Endothelins secreted from human keratinocytes are intrinsic mitogens for human melanocytes. J Biol Chem 1992;267:24675–80.
Contents lists available at SciVerse ScienceDirect
Cytokine journal homepage: www.elsevier.com/locate/issn/10434666
Paracrine cytokine interaction between UVB-exposed epidermal keratinocytes and dermal fibroblasts in stimulating expression of skin fibroblast-derived elastase Hiroaki Nakajima, Ryosuke Yoshioka, Yoshiyuki Ezaki, Tomoyashu Nagai, Genji Imokawa ⇑ School of Bioscience and Biotechnology, Tokyo University of Technology, Japan
a r t i c l e
i n f o
Article history: Received 25 November 2011 Received in revised form 24 February 2012 Accepted 3 March 2012 Available online 14 April 2012 Keywords: Ultraviolet light Elastase Wrinkling Skin fibroblasts Neutral endopeptidase
a b s t r a c t Background: We recently reported that over-expression of skin fibroblast-derived elastase (SFE) plays a pivotal role in the mechanism of UVB-induced skin wrinkling. Since UVB penetrates only modestly to the dermis, we hypothesized that factors secreted by UVB-exposed keratinocytes in the epidermis trigger fibroblasts in the dermis to increase their expression of SFE which then degrades the elastic fibers. Objective: In this study, we characterized the paracrine interaction between human keratinocytes (HK) and human fibroblasts (HF) which leads to increased expression of SFE. Methods and results: Medium conditioned by UVB-exposed HK contained increased levels of IL-1a, GMCSF, IL-6, TNFa and IL-8. While HF cultured with those conditioned medium slightly down-regulated the gene expression of collagen and elastin, they significantly increased their expression of SFE at the transcriptional, translational and enzymatic levels. Neutralizing antibodies to IL-1a or GM-CSF significantly abolished the increased expression of SFE at the translational and/or enzymatic levels in HF cultured with those conditioned medium, while neutralizing antibodies to IL-6, IL-8 or TNFa had no such effect. The addition of IL-1a or GM-CSF, but not TNFa, IL-6 or IL-8, at concentrations ranging from 1 to 10 nm, significantly stimulated the enzymatic levels of SFE in HF. Conclusions: The sum of these findings suggests that IL-1a and GM-CSF are intrinsic cytokines secreted by UVB-exposed HK that stimulate expression of SFE by HF, leading to UVB-induced wrinkle formation. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Repetitive exposure to sunlight has been thought to be the most plausible factor that causes skin wrinkles since they frequently appear on the face. However, little is known about the precise physicochemical and biological mechanism(s) which lead to the formation of wrinkles due to sunlight exposure. In clinical studies, the formation of facial wrinkles has been associated with the loss of elastic properties of the skin [1,2]. Repetitive exposure of the skin to UVB radiation at suberythemal doses significantly reduces its elastic properties, which in turn leads to the formation of wrinkles [3]. Repetitive UVB irradiation of the skin also elicits a marked alteration in the microstructure of collagen fibers [4–6] as well as the three-dimensional structure of elastic fibers, which are closely linked to a subsequent reduction in elastic properties [7,8]. The Abbreviations: GM-CSF, granulocyte macrophage colony stimulatory factor; IL1a, Interleukin-1a; NEP, neutral endopeptidase; SFE, skin fibroblast-derived elastase; STANA, N-succinyl-tri-alanyl-p-nitroaniline; UVB, ultraviolet B; MMP-1, matrix-meltalloproteinase-1. ⇑ Corresponding author. Address: School of Bioscience and Biotechnology, Tokyo University of Technology, 1404-1 Katakura, Hachioji, Tokyo 192-0982, Japan. Tel./fax: +81 42 637 2424. E-mail address: [email protected] (G. Imokawa). 1043-4666/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.cyto.2012.03.008
alteration of collagen bundles and elastic fiber architecture has been substantiated by a marked continuous up-regulation of a collagen- or an elastin-degrading enzymes, collagenase-I (matrixmetalloproteinase (MMP)-I or elastase in wrinkled skin after repetitive UVB irradiation [9–12]as well as after ovariectomy [13]. A study to determine which matrix protease(s) are linked to the loss of skin elasticity in the dermis of UVB-irradiated mouse skin reported a significant increase in the activity of elastase in UVBirradiated skin compared with non-irradiated skin [13]. In contrast, there was a slight but not significant increase in the activity of collagenase I in the dermis at week 14 in UVB-irradiated skin compared with non-irradiated skin [13]. On the other hand, there was a decrease in the activity of collagenase IV in UVB-irradiated skin at the same time point compared with non-irradiated skin [13]. On the other hand, UVA-irradiation elicited skin sagging like age-related morphological structure of the human cheek [14] rather than wrinkling as a result of higher up-regulation of MMP-1 expression compared with skin elastase expression in the dermis [13] and in cultured human fibroblasts [15]. Thus, we hypothesized that enhanced elastase activity in the dermis of UVB-irradiated skin plays an important role in the degeneration of elastic fibers, which eventually results in the loss of skin elasticity. This elastase-triggered alteration of the three-dimensional
H. Nakajima et al. / Cytokine 59 (2012) 166–175
structure of elastic fibers was validated by our UVB studies [16,17] which used a synthetic inhibitor specific for skin fibroblast elastases to demonstrate a close interrelationship among wrinkle formation, elastic properties and elastic fiber linearity, a marker for the alteration of elastic fibers. The fact that elastase inhibitors can serve as anti-wrinkling agents was also substantiated by another animal study [18] and by 1 year clinical study of human volunteers [19]. The sum of this evidence strongly suggests that enhanced elastase activity produced by dermal fibroblasts plays a pivotal role in the mechanism of UVB-induced wrinkling of the skin. Skin fibroblast-derived elastase (SFE) is a 94 kDa membranebound type metalloprotease with a neutral optimum pH [20–23]. Although there were several candidate enzymes for SFE, such as 92, 72 kDa type IV collagenase (gelatinase) [24,25], neutrophil elastase [26], cathepsin G [27] and protease 3 [28], none of those had properties which matched the characteristics of SFE. Since there are similarities between SFE and neutral endopeptidase (NEP) [29] in terms of their size (MW 97,000), their membranebound nature, and their inhibitory profiles, we used immunoprecipitation and Western blotting with an antibody to NEP to show that SFE is identical to NEP (here after referred as to SFE) [30]. The identification of SFE as NEP enabled us to clarify epithelialmesenchymal interactions, especially at the gene and protein levels, whereby UVB irradiation that penetrates only the epidermis [31] causes fibroblasts in the dermis to stimulate their expression of NEP. Thus, we hypothesized that soluble cytokines secreted by UVB-exposed keratinocytes pass through the basement membrane into the dermis and cause dermal fibroblasts to increase their expression of NEP which then disrupts the three dimensional networks of elastic fibers. This results in the loss of skin elastic properties, an essential parameter that leads to wrinkle formation. In this study, we characterized the paracrine linkage between UVB-exposed epidermal keratinocytes and dermal fibroblasts that leads to wrinkle formation in the skin. We determined factors involved in the up-regulation of fibroblast elastases by examining the stimulatory effects of conditioned medium from UVB-exposed human keratinocytes on fibroblast elastases and the inhibitory effects of neutralizing antibodies to various paracrine factors. We report that interleukin-1a and granulocyte macrophage colony stimulatory factor are intrinsic cytokines secreted by UVB-exposed human keratinocytes that stimulate the expression of NEP by human fibroblasts.
167
2. Materials and methods 2.1. Materials Human primary keratinocytes (HPK) (derived from adult abdomen skin, age 32, female, Caucasian), normal human fibroblasts (HF) (derived from foreskin, Caucasian), serum-free keratinocyte growth medium (Medium 154S) containing low calcium (0.2 mM), bovine pituitary extract (BPE), and epidermal growth factor (EGF), were obtained from Kurabo (Tokyo, Japan). Human keratinocyte cell line, HaCaT cells were kindly supplied by Dr. M. Furue (Kyushu University, School of Medicine, Department of Dermatology). Anti-human NEP (CD10:SN5C: sc-19586) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Neutralizing antibodies such as anti-human IL-6 monoclonal mouse antibody (MAB206), anti-human CXCL8/IL-8 monoclonal mouse antibody (MAB208), anti-human TNF-a monoclonal mouse antibody (MAB2101) and anti-human GM-CSF monoclonal mouse antibody (MAB215) were purchased from R&D Systems (Minneapolis, MN, USA). The neutralizing mouse antibody to human IL-1a (hIL-1a)(L122M) and normal mouse IgG (12-371) was obtained from EXALPHA Biologicals (Shirley, MA, USA) and Millipore Co. Ltd., respectively. Recombinant human IL-6 (206-IL-010), recombinant human CXCL8/IL-8 (618-IL-010) and recombinant human GM-CSF (215-GM-010) were purchased from R&D Systems. Recombinant human IL-1a (200-01A) and recombinant human TNFa (300-01A) were obtained from PEPROTECH (London, UK). ELISA Kits (for IL-1a, IL-6, IL-8, GM-CSF and TNFa) were obtained from Endogen (Thermo Fisher Scientific, Yokohama, Japan). Horseradish peroxidase-conjugated goat polyclonal anti-mouse IgG was obtained from Transduction Laboratories (Franklin Lakes, NJ, USA). Other chemicals of reagent grade were purchased from Sigma-Aldrich (St. Louis, MO, USA). 2.2. Cell cultures HF were cultivated as previously described [15]. HPK were maintained in serum-free keratinocyte medium (Medium 154S) (Kurabo, Tokyo Japan) supplemented with 5 ng/ml EGF and 50 lg/ml BPE at 37 °C with 5% CO2. HaCaT cells were maintained in Dulbecco modified Eagle’s medium (DMEM) with 10% FCS, 100 lg/ml penicillin, 100 lg/ml streptomycin, and 250 ng/ml
Fig. 1. Effects of conditioned medium from UVB-exposed HPK on mRNA levels of NEP, elastin, collagenase I and collagen in HF as revealed by real-time RT-PCR analysis. The conditioned medium from UVB-exposed HPK was added at 1:1 dilution to cultured HF to measure expression levels of each mRNA at 4 h post-incubation. Values are means ± S.D. derived from three independent experiments. ⁄p < 0.05. ⁄⁄p < 0.01.
168
H. Nakajima et al. / Cytokine 59 (2012) 166–175
Fig. 2. Effects of conditioned medium from UVB-exposed HPK or HaCaT cells on NEP protein (A) and enzymatic levels (B) in HF. (A): NEP protein levels were measured by Western blotting analysis 24, 48 and 72 h after the addition of conditioned medium (CM) from UVB (0 or 80 mJ/cm2)-exposed HaCaT cells or non-conditioned fresh DMEM medium (NCM) at 1:1 dilution to cultured HF. Representative immunoblots from three independent experiments are shown. (B): (a) HPK; (b) HaCaT cells. NEP activity levels were measured by enzymatic assay using STANA 72 h after the addition of the condition medium at 1:1 dilution to cultured HF. Values are means ± S.D. derived from three independent experiments. ⁄p < 0.05. ⁄⁄p < 0.01.
amphotericin B at 37 °C in a 95% air, 5% CO2 atmosphere. In experiments to measure cytokine levels by ELISA kits and to collect conditioned medium after UVB irradiation, HPK or HaCaT cells were cultured in 10 cm dishes with 5 ml of Medium 154S without EGF and BPE or with 5 ml of DMEM without FCS, respectively. For experiments to test paracrine interactions, the keratinocyte-conditioned medium was obtained by culturing HPK in 10 cm dishes with 5 ml of Medium 154S without EGF and BPE or HaCaT cells in 10 cm dishes with 5 ml of DMEM without FCS for 3 days after UVB irradiation. HF were seeded in 6 cm dishes at a density of approximately 5 105 cells with DMEM with 10% FCS and cultured
for 24 h. After 24 h starvation in DMEM without FCS, 1.0 ml of the medium conditioned by UVB exposed keratinocytes was added at 1:1 dilution to 1.0 ml of fresh culture medium (DMEM) of HF. 2.3. UVB irradiation HPK or HaCaT cells cultured at approximately 1.5–2.0 106 cells in 10 cm dishes were washed twice with PBS and were then exposed to UVB. The UVB source was a SE fluorescent lamp (Clinical Supply, Tokyo, Japan) that emitted an energy spectrum with high fluency in the UVB region (280–320 nm) with a peak at
H. Nakajima et al. / Cytokine 59 (2012) 166–175
169
Fig. 3. Secretion levels of cytokines (pg/ml) in the medium from HaCaT cells or HPK 72 h after UVB irradiation at 0–80 mJ/cm2. (a) HaCaT cells; (b) HPK. Cultured HaCaT cells or HPK were exposed to UVB at the indicated doses and secretion levels of cytokines were measured by ELISA kit at 72 post-irradiation. Values are means ± S.D. derived from three independent experiments. ⁄⁄p < 0.01. ⁄p < 0.05 (vs. 0 mJ/cm2).
305 nm. The emitted dose was calculated using a UVB radiometer photodetector (Torex, Tokyo, Japan). After UVB irradiation, the cells were maintained for 48–72 h in 5 ml of DMEM without FCS (for HaCaT cells) or in 5 ml of Medium 154S without EGF and BPE (for HPK) at 37 °C in a 95% air, 5% CO2 atmosphere. Preliminary testing determined no significant effect on cell viability of HPK and HaCaT cells by UVB irradiation with 80 mJ/cm2 at 48 and 72 h post-irradiation as measured by MTT assay [32]. 2.4. Neutralizing antibody experiments After 1.0 ml of the conditioned medium obtained from UVB- or mock-exposed HaCaT cells or HPK were added to 1.0 ml of fresh DMEM, neutralizing antibodies diluted in PBS were added at the indicated concentrations into the diluted conditioned medium. After replaced with the prepared conditioned medium, HF were maintained to culture for 72 h in 6 cm dishes. 2.5. Measurement of elastase activity Fibroblast enzyme solution was prepared as previously described [15]. Elastase activity using the synthetic substrate N-succinyl-tri-alanyl-p-nitroaniline (STANA) (Peptide Institute Inc., Osaka, Japan) was measured as previously described [33]. The release of p-nitroaniline was measured by absorbance at 405 nm and enzymatic activity is expressed as nmol p-nitroaniline/h/lg protein. Finally a comparison was performed by expressing activity as relative to elastase activity.
to X-ray film for the specified times to detect bands, and specific bands were quantitated by densitometric scanning and were analyzed with Quantity OneÒ (Bio Rad). 2.7. Real-time RT-PCR analysis Levels of NEP, MMP-1, collagen and elastin mRNAs in HF were examined using real-time RT-PCR as previously described [15]. Primers with the following sequences were used: GAPDH: Forward, 50 -GAAGGTGAAGGTCGGAGTCAACG-30 , Reverse, 50 -AGTCCTTCCACGATAACCAAAGTTG-30 384 bp, NEP: Forward, 50 -GAACCTACAAGGAGTCCAGA-30 Reverse, 50 -GCTCCACTTATCCACTCATC-30 121 bp, Collagenase I (MMP-1): Forward, 50 -GCTGGGAGCAAACACATCTGAGGT-30 , Reverse, 50 -TGAGCCGCAACACGATGTAAGTTG-30 331 bp, Elastin: Forward, 50 -AGGTGTATACCCAGGTGGCGTGCT-30 , Reverse, 50 -CAACCCCTGTCCCTGTTGGGTAAC-30 311 bp, Collagen: Forward, 50 -CTGGTCCCCAAGGCTTCCAAGGTC-30 , Reverse, 50 -CCATCATTTCCACGAGCACCAGCA-30 391 bp. Preliminary testing determined no significant effect on GAPDH levels in HPK and HaCaT cells by UVB irradiation with 80 mJ/cm2 at 4–24 h post-irradiation. 2.8. Statistical analysis All data are expressed as means ± SD (n = 3). For pairwise comparisons, either Student’s t-test or Welch’s t-test was applied. For multiple comparisons, data were tested by one-way ANOVA, and subsequently using the Tukey or Dunnet multiple comparison test. P values less than 0.05 are considered statistically significant. 3. Results
2.6. Western blotting Cell lysates were separated on 10% SDS-PAGE gels, and were blotted to PVDF membranes (Bio-Rad, Hercules, CA, USA). After blocking with 3% bovine serum albumin-containing Tris/HCl (pH 7.5, 100 mM NaCl), the membranes were treated with the mouse monoclonal anti-human NEP antibody at room temperature for 1 h, then with anti-mouse IgG conjugated to horseradish peroxidase, and finally with Amersham ECL reagents. Plots were exposed
3.1. Effects of conditioned medium from UVB-exposed human primary keratinocytes on mRNA levels in human fibroblasts To test the paracrine cytokine mechanism underlying UVB-induced wrinkle formation, we asked initially whether conditioned medium from UVB-exposed human primary keratinocytes (HPK) would stimulate the expression of genes encoding matrix proteins or matrix metalloproteases in human fibroblasts (HF). Real-time
170
H. Nakajima et al. / Cytokine 59 (2012) 166–175
Fig. 4. Neutralizing antibody abrogates the stimulated NEP activity. (A): HaCaT cells, (a) anti-human IL-1a antibody, (b) anti-human GM-CSF antibody, (c) anti-human IL-6 antibody, (d) anti-human IL-8 antibody, (e) anti-human TNFa antibody, (f)non-specific mouse IgG. (B): HPK, (a) anti-human IL-1a antibody, (b) anti-human GM-CSF antibody. Antibodies as well as non-specific mouse IgG were incubated at the indicated concentrations with the two fold diluted conditioned medium. After replaced with the prepared conditioned medium, HF were maintained to culture for 72 h. Lysates of HF were then assessed for NEP activity using STANA. Values are means ± S.D. derived from three independent experiments. ⁄⁄p < 0.01. ⁄p < 0.05.
RT-PCR for NEP mRNA revealed that when the HPK-conditioned medium was added at 1:1 dilution to cultured HF, levels of NEP mRNA were significantly stimulated at 4 post-incubation with a UVB dose of 80 mJ/cm2 compared with mock-irradiated controls (Fig. 1). Real-time RT-PCR for elastin mRNA revealed that the conditioned medium from UVB-exposed HPK slightly down-regulated elastin mRNA levels at 4 h post-incubation with UVB doses of 60 or 80 mJ/cm2 (Fig. 1). Real-time RT-PCR for collagenase type I mRNA revealed that the conditioned medium from UVB-irradiated HPK significantly stimulated collagenase type I (MMP-1) mRNA levels at 4 h post-incubation with a UVB dose of 80 mJ/cm2 (Fig. 1). Real-time RT-PCR for collagen mRNA revealed that the conditioned medium from UVB-irradiated HPKslightly down-regulated collagen mRNA levels at 4 h post-incubation with UVB doses of 60 or 80 mJ/cm2 (Fig. 1). 3.2. Effects of conditioned medium from UVB-exposed HaCaT cells on NEP protein levels in HF Since mRNA expression levels for NEP in HF were distinctly stimulated by the conditioned medium from UVB-exposed human keratinocytes, we next asked whether the conditioned medium stimulates NEP protein levels in HF. Western blotting analysis for
NEP protein revealed that when the conditioned medium from UVB-exposed HaCaT cells at a dose of 80 mJ/cm2 was added at 1:1 dilution to cultured HF, NEP protein levels were markedly stimulated at 72 h post-incubation as compared with the mockirradiation controls (Fig. 2A). 3.3. Effects of conditioned medium from UVB-exposed human keratinocytes on NEP activity in human fibroblasts Since the NEP mRNA and protein levels in HF were significantly stimulated by the conditioned medium from UVB-exposed human keratinocytes, we next asked whether the conditioned medium stimulates the enzymatic activity of NEP in HF. Enzymatic assays for NEP revealed that when the conditioned medium obtained from UVB-exposed HaCaT cells and from UVB-exposed HPK was added at 1:1 dilution to cultured HF, the enzymatic activity of NEP were significantly stimulated at 72 h post-incubation as compared with the mock-irradiation controls (Fig. 2B). 3.4. Cytokine secretion in UVB-exposed human keratinocytes When cultured HPK or HaCaT cells were exposed to UVB at the doses of 0–80 mJ/cm2 and cytokines secreted into the conditioned
H. Nakajima et al. / Cytokine 59 (2012) 166–175
171
Fig. 4 (continued)
medium were measured by ELISA kits, IL-1a, IL-6, IL-8, and GMCSF were significantly increased at 72 h post-irradiation (Fig. 3).
3.7. Effects of cytokines on the enzymatic activity of NEP in human fibroblasts
3.5. Neutralizing antibody abrogates the stimulated NEP activity
We next examined the potential of secreted cytokines to stimulate the enzymatic activity of NEP in HF. When HF were treated with IL-1a, GM-CSF, IL-6 or IL-8 at 0, 1, 5, and 10 nm, the enzymatic activity of NEP was significantly stimulated over 72 h by incubation with IL-1a or GM-CSF, but not by IL-6, IL-8 or TNFa (Fig. 6).
Since HF cultured with the conditioned medium from UVB-exposed HPK significantly increased the expression of NEP at the transcriptional, translational and enzymatic levels, we have attempted to identify kinds of cytokines mainly responsible for the stimulated expression of NEP by neutralizing experiments with antibodies to cytokines secreted in the conditioned medium. When conditioned medium obtained from UVB- or non-exposed HaCaT cells was treated with neutralizing antibodies to IL-1a, GM-CSF, IL-6, TNFa or IL-8 or with a non-specific mouse IgG, the stimulation of NEP activity was markedly abrogated by anti-IL-1a and anti-GM-CSF antibodies but not by anti-IL-6, anti-IL-8 or anti-TNFa or the non-specific mouse IgG (Fig. 4A). When conditioned medium obtained from UVB- or non-exposed HPK was treated with anti-IL-1a or anti GM-CSF antibodies, the stimulation of NEP activity was markedly abrogated for UVB-exposed HPK but not for non-exposed cells by anti-IL-1a and by anti-GM-CSF antibodies (Fig. 4B). 3.6. Neutralizing antibody abrogates the stimulated NEP protein When conditioned medium obtained from UVB- or non-exposed HaCaT cells was treated with IL-1a, GM-CSF, IL-6, or IL-8 antibodies or with non-specific mouse IgG, the increased level of NEP protein was significantly abrogated by IL-1a and GM-CSF antibodies, but not by IL-6, IL-8, TNFa antibodies or by a non-specific mouse IgG (Fig. 5).
4. Discussion Although the enhanced activity of NEP in skin fibroblasts is responsible for the UVB-induced cascade of biological events that leads to the degeneration of elastic fibers and the reduced skin elasticity, which in turn leads to wrinkle formation [17], little is known about the mechanism(s) whereby NEP activity in dermal fibroblasts is stimulated by UVB irradiation. Based on the fact that UVB penetrates to the dermis to only a minor extent [31] and that UVB down-regulates NEP expression in HF (unpublished data), the most plausible mechanism is that UVB elicits the production of a factor(s) in the epidermis which then stimulates the expression of elastases by fibroblasts in the dermis. To test that hypothesis, we asked whether conditioned medium from UVB-exposed human keratinocytes stimulates the expression of matrix proteins or matrix metalloproteases, especially NEP, in HF. Real-time RT-PCR analysis revealed that the conditioned medium from UVB-exposed human keratinocytes stimulates NEP mRNA levels and collagenase type I mRNA levels. In contrast, levels of mRNAs encoding elastin as well as collagen are down-regulated at 4 h post-incubation.
172
H. Nakajima et al. / Cytokine 59 (2012) 166–175
Fig. 5. Neutralizing antibody abrogates the stimulated NEP protein. (A): anti-human IL-1a antibody, (B): anti-human GM-CSF antibody, (C): anti-human IL-6 antibody, (D): anti-human IL-8 antibody, (E): anti-human TNFa antibody, (F): non-specific mouse IgG. Antibodies as well as non-specific mouse IgG were added at the indicated concentrations to the two fold diluted conditioned medium. After replaced with the prepared conditioned medium, HF were maintained to culture for 72 h. Lysates of HF were then subjected to Western blotting analysis. Values are means ± S.D. derived from three independent experiments. ⁄⁄p < 0.01. ⁄p < 0.05.
Western blotting and enzymatic assay of HF treated with conditioned medium from UVB-exposed human keratinocytes demonstrated that NEP protein and enzymatic levels in human fibroblasts are significantly increased at 48 or 72 h post-incubation. The sum of these findings indicates that paracrine factors are secreted by UVB-exposed human keratinocytes which penetrate into the dermis to trigger dermal fibroblasts to stimulate NEP activity. Comparison of the in vivo profiles for matrix
proteins/metallo-proteinases in UVB-exposed and wrinkled skin and the in vitro cellular effects of conditioned medium from UVB-exposed human keratinocytes on gene expression patterns in HF (in both of which, while there is a down-regulated expression of elastin mRNA, NEP and MMP-1mRNAs are markedly up-regulated) [17] revealed that the effects of the conditioned medium on HF mimics the in vivo situation in wrinkled skin despite the fact that the frequency of UVB exposure differs greatly between the
H. Nakajima et al. / Cytokine 59 (2012) 166–175
173
Fig. 6. Effects of cytokines on the enzymatic activity of SFE HF as revealed by enzymatic assay using STANA. NEP activity levels were measured by enzymatic assay using STANA 72 h after the addition of cytokines to cultured HF at the indicated concentrations. Values are means ± S.D. derived from three independent experiments. IL-1a (a), GM-CSF (b), IL-6 (c), IL-8 (d), TNFa (e). ⁄⁄p < 0.01.
Fig. 7. A scheme for wrinkling mechanism. Repetitive UVB exposure causes keratinocytes to secrete IL-1a which triggers GM-CSF secretion in an autocrine fashion. The secreted IL-1a and GM-CSF penetrate the dermis to stimulate the expression of NEP by fibroblasts, which then cleaves elastic fibers surrounding the fibroblasts, leading to deterioration of the three-dimensional configuration of elastic fibers. This results in a loss of skin elasticity and in turn leads to wrinkle formation.
174
H. Nakajima et al. / Cytokine 59 (2012) 166–175
in vitro and in vivo situations. This suggested that the enhanced activity of NEP or MMP-1 in UVB-irradiated skin [12,13] is mediated by basement membrane-permeable soluble factors secreted by UVB-exposed human keratinocytes. Taken together, the sum of available evidence supports our hypothesis for a mechanism of wrinkle formation by which cytokines are released by keratinocytes following UVB irradiation, triggering dermal fibroblasts to stimulate the expression and activity of NEP. The enhanced NEP activity then results in the deterioration of the three-dimensional architecture of elastic fibers, reducing skin elasticity, and eventually leading to the formation of wrinkles. To determine which cytokines secreted by UVB-exposed human keratinocytes are intrinsic soluble factors that cause HF to stimulate the activity of NEP, we used conditioned medium to evaluate the inhibitory effects of several neutralizing antibodies to cytokines or a chemokine secreted by UVB-exposed human keratinocytes on NEP expression in HF. Those results revealed that among the neutralizing antibodies tested, the anti-IL-1a and anti-GM-CSF (but not anti-IL-6, anti-IL-8 or anti-TNFa or the non-specific mouse IgG) significantly abolished the enhanced activity of NEP in HF after the addition of the conditioned medium of UVB-exposed human keratinocytes. The neutralizing effects of the anti-IL-1a and the anti-GM-CSF antibodies on enzymatic activity levels were accompanied by the down-regulated protein expression of NEP. The sum of these findings strongly suggests that IL-1a and GM-CSF are intrinsic keratinocyte-derived cytokines associated with the up-regulated expression of NEP. Interestingly, those two antibodies also reduced the activity of NEP after the addition of the conditioned medium from non-exposed HaCaT cells. These findings indicate that the conditioned medium from non-exposed HaCaT cells also contains IL-1a and GM-CSF at concentrations sufficient, although to a lesser extent, to stimulate NEP activity in HF. Before speculating about the paracrine cytokine mechanism involved in the activation of NEP in vivo, it should be noted that studies using conditioned medium have several circumstances distinct from the in vivo situation; (1) The conditioned medium used human keratinocytes exposed to single UVB in contrast to the repeated UVB exposure under in vivo conditions required for wrinkle formation [11,12]. (2) The conditioned medium contains accumulated levels of cytokines that were secreted for 72 h postirradiation, which differs from the in vivo situation where shortly after UVB irradiation, secreted cytokines soon penetrate the epidermis to stimulate fibroblasts in the dermis, leading to the enhanced expression of NEP. (3) Since several cytokines secreted by UVB-exposed keratinocytes result from autocrine loop stimulation by primarily released cytokines such as IL-1a [34], the concentration of each cytokine in the conditioned medium may differ, which may not precisely reflect the repeated and accumulated secretion of NEP-inducing cytokines under the in vivo long term conditions required for wrinkle formation. In spite of these reservations, our data clearly show that IL-1a and GM-CSF are the only cytokines whose antibodies have specific neutralizing effects on the stimulated activity or protein level of NEP in HF cultured with the conditioned medium from UVB-exposed human keratinocytes. Consistent with this, they are found to be the cytokines capable of significantly stimulating the enzymatic activity in HF although it remains to be clarified whether IL-1a serves in an autocrine loop to trigger the secretion of GM-CSF and/or could directly cause skin fibroblasts to stimulate the activity of NEP.
5. Conclusion Thus, as depicted in Fig. 7, we propose a UVB-induced wrinkling mechanism as follows: Repetitive UVB exposure causes keratino-
cytes to secrete IL-1a which triggers GM-CSF secretion in an autocrine fashion. The secreted IL-1a and GM-CSF penetrate the dermis to stimulate the expression of NEP by fibroblasts, which then cleaves elastic fibers surrounding the fibroblasts [23], leading to deterioration of the three-dimensional configuration of elastic fibers. This results in a loss of skin elasticity and in turn leads to wrinkle formation. Conflict of interest The authors have declared that no conflict of interest exists. References [1] Akasaki S, Imokawa G. Mechanical methods for evaluating skin surface architecture in relation to wrinkling. J Dermatol Sci 2001;27(Suppl. 1):S5–S10. [2] Akazaki S, Nakagawa H, Kazama H, Osanai O, Kawai M, Takema Y, et al. Agerelated changes in skin wrinkles assessed by a novel three-dimensional morphometric analysis. Br J Dermatol 2002;147:689–95. [3] Takema Y, Imokawa G. The effects of UV-A and UV-B irradiation on the viscoelastic properties of hairless mouse skin in vivo. Dermatology 1998;196:397–400. [4] Schwartz E. Connective tissue alterations in the skin of ultraviolet irradiated hairless mice. J Invest Dermatol 1988;91:158–61. [5] Zheng P, Kligman LH. UVA-induced ultrastructural changes in hairless mouse skin: a comparison to UVB-induced damage. J Invest Dermatol 1993;100:194–9. [6] Rabe JH, Mamelak AJ, McElgunn PJS, Morison WL, Sauder DN. Photoaging: mechanisms and repair. J Am Acad Dermatol 2006;55:1–19. [7] Imayama S, Nakamura K, Takeuchi M, Hori Y, Takema Y, Sakaino Y, et al. Ultraviolet-B irradiation deforms the configuration of elastic fibers during the induction of actinic elastosis in rats. J Dermatol Sci 1994;7:32–8. [8] Imokawa G, Takema Y, Yorimoto Y, Tsukahara K, Kawai M, Imayama S. Degree of ultraviolet-induced tortuosity of elastic fibers in rat skin is age-dependent. J Invest Dermatol 1995;105:254–8. [9] Fisher GJ, Datta SC, Talwar HS, Wang ZQ, Varani J, Kang S, et al. Molecular basis of sun-induced premature skin ageing and retinoid antagonism. Nature 1996;379:335–9. [10] Brennan M, Bhatti H, Nerusu KC, Bhagavathula N, Kang S, Fisher GJ, et al. Matrix metalloproteinase-1 is the major collagenolytic enzyme responsible for collagen damage in UV-irradiated human skin. Photochem Photobiol 2003;78:43–8. [11] Tsukahara K, Takema Y, Moriwaki S, Tsuji N, Suzuki Y, Fujimura T, et al. Selective inhibition of skin fibroblast elastase elicits a dose-dependent prevention of UVB-induced wrinkle formation. J Invest Dermatol 2001;117:671–7. [12] Tsuji N, Moriwaki S, Suzuki Y, Takema Y, Imokawa G. The role of elastases secreted by fibroblasts in wrinkle formation: implication through selective inhibition of elastase activity. Photochem Photobiol 2001;74:283–90. [13] Tsukahara K, Nakagawa H, Moriwaki S, Shingo K, Ohuchi A, Takema Y, et al. Ovariectomy is sufficient to accelerate spontaneous skin ageing and to stimulate ultraviolet irradiation-induced photoageing of murine skin. Br J Dermatol 2004;151:984–94. [14] Tsukahara K, Takema Y, Fujimura T, Moriwaki S, Kitahara T, Imokawa G. Determination of age-related changes in the morphological structure (sagging) of the human cheek using a photonumeric scale and three-dimensional surface parameters. Int J Cosmet Sci 2000;22:247–58. [15] Suganuma K, Nakajima H, Otsuki M, Imokawa G. Astaxanthin attenuates the UVA-induced up-regulation of matrix-metalloproteinase-1 and skin fibroblast elastase in human dermal fibroblasts. J Dermatol Sci 2010;58:136–42. [16] Imokawa G. Recent advances in characterizing biological mechanisms underlying UV-induced wrinkles: a pivotal role of fibrobrast-derived elastase. Arch Dermatol Res 2008;300(Suppl. 1):S7–S20. [17] Imokawa G. Mechanism of UVB-induced wrinkling of the skin: paracrine cytokine linkage between keratinocytes and fibroblasts leading to the stimulation of elastase. J Invest Dermatol Symp Proc 2009;14:36. [18] Tsukahara K, Nakagawa H, Moriwaki S, Takema Y, Fujimura T, Imokawa G. Inhibition of ultraviolet-B-induced wrinkle formation by an elastase-inhibiting herbal extract: implication for the mechanism underlying elastase-associated wrinkles. Int J Dermatol 2006;45:460–8. [19] Tsukahara K, Nakagawa H, Ohtuki M, Moriwaki S, Fujimura T, Takema Y, et al. A clinical effect of Zingiber officinale (L.) rose on facial wrinkles and skin elasticity. J Mol Biol 2004;27:19–27. [20] Godeau G, Hornebeck W. Morphometric analysis of the degradation of human skin elastic fibers by human leukocyte elastase (EC 3–4-21-37) and human skin fibroblast elastase (EC 3–4-24). Pathol Biol 1988;36:1133–8. [21] Homsy R, Pelletier-Lebon P, Tixier JM, Godeau G, Robert L, Hornebeck W. Characterization of human skin fibroblast elastase activity. J Invest Dermatol 1988;91:472–7. [22] Croute F, Delaporte E, Bonnefoy JY, Fertin C, Thivolet J. Nicolas JF Interleukin-1 beta stimulates fibroblast elastase activity. Br J Dermatol 1991;124:538–41.
H. Nakajima et al. / Cytokine 59 (2012) 166–175 [23] Mecham RP, Broekelmann TJ, Fliszar CJ, Shapiro SD, Welgus HG, Senior RM. Elastin degradation by matrix metalloproteinases. J Biol Chem 1997;272:18071–6. [24] Senior RM, Griffin GL, Fliszar CJ, Shapiro SD, Goldberg GI, Wegus HG. Human 92- and 72-kilodalton type IV collagenase are elastase. J Biol Chem 1991;266:7870–5. [25] Shipley JM, Doyle GAR, Fliszar CJ, Ye Q-Z, Johnson LL, Shapiro SD, et al. The structural basis for the elastolytic activity of the 92-kDa and 72-kDa gelatinases: role of the fibronectin type II-like repeats. J Biol Chem 1996;271:4335–41. [26] Cambell EJ, Cambell MA. Pericellular proteolysis by neutrophils in the presence of proteinase inhibitors: effects of substrate opsonization. J Cell Biol 1988;106:667. [27] Boudier C, Godeau G, Hornebeck W, Robert L, Bieth JG. The elastolytic activity of cathepsin G: an ex vivo study with dermal elastin. Am J Respir Cell Mol Biol 1991;4:497. [28] Kao RC, Wehner NG, Skubitz KM, Gray BH, Hoidal JR. Proteinase 3. A distinct human polymorphonuclear leukocyte proteinase that produces emphysema in hamsters. J Clin Invest 1988;82:1963–73.
175
[29] Fulcher IS, Matsas R, Turner AJ, Kenny AJ. Kidney neutral endopeptidase and the hydrolysis of enkephalin by synaptic membranes show similar sensitivity to inhibitors. Biochem J 1982;203:519–22. [30] Morisaki N, Moriwaki S, Sugiyama-Nakagiri Y, Haketa K, Takema Y, Imokawa G. Neprilysin is identical to skin fibroblast elastase – its role in skin ageing and UV responses. J Biol Chem 2010;285:39819–27. [31] Bruls WA, Slaper H, van der Leun JC, Berrens L. Transmission of human epidermis and stratum corneum as a function of thickness in the ultraviolet and visible wavelengths. Photochem Photobiol 1984;40:485–94. [32] Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63. [33] Nakagawa K, Tsuji T, Kadoya A, Hamada T. Elastase-like enzyme activity in cultured human fibroblast. Skin Re 1987;29:793–7. [34] Imokawa G, Yada Y, Miyagishi M. Endothelins secreted from human keratinocytes are intrinsic mitogens for human melanocytes. J Biol Chem 1992;267:24675–80.