If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Molecular Biochemistry, Gunma University Graduate School of Medicine, Maebashi, Gunma, JapanDepartment of Dermatology, Gunma University Graduate School of Medicine, Maebashi, Gunma, Japan
Department of Molecular Biochemistry, Gunma University Graduate School of Medicine, Maebashi, Gunma, JapanDepartment of Dermatology, Gunma University Graduate School of Medicine, Maebashi, Gunma, Japan
G2A is a stress-inducible G protein-coupled receptor for oxidized free fatty acids, such as 9-hydroxyoctadecadienoic acid (HODE). As skin is routinely and pathologically exposed to many oxidative stresses such as UV radiation, chemical agents, and inflammation that might induce both G2A expression and production of G2A ligands, we examined G2A function in human keratinocytes. G2A was expressed in human epidermis, normal human epidermal keratinocytes (NHEK), and an immortalized human keratinocyte cell line (HaCaT). 9(S)-HODE evoked intracellular calcium mobilization and secretion of cytokines, including IL-6, IL-8, and GM-CSF in NHEK cells. These responses became prominent in HaCaT cells by overexpression of G2A. 9(S)-HODE inhibited proliferation of NHEK cells by suppressing DNA synthesis and arresting the cell cycle in the G0/1-phase. On the other hand, 13(S)-HODE, another major oxidative product from linoleate, showed little or no effect on either cytokine secretion or on proliferation in NHEK cells. A small interfering RNA designed to downregulate G2A caused suppression of 9(S)-HODE-induced inhibitory effects on proliferation of NHEK cells. UVB and H2O2 induced G2A expression and caused oxidation of linoleate to produce 9-HODE in HaCaT cells. These results suggest that 9-HODE-G2A signaling plays proinflammatory roles in skin under oxidative conditions.
Abbreviations:
GPCR
G protein-coupled receptor
HODE
hydroxyoctadecadienoic acid
NHEK
normal human epidermal keratinocyte
PBS
phosphate-buffered saline
PPAR
peroxisome proliferator-activated receptor
ROS
reactive oxygen species
siRNA
small interfering RNA
RT
reverse transcriptase
Introduction
G2A (derived from G2 accumulation) was first identified as a stress-inducible G protein-coupled receptor (GPCR) predominantly expressed in lymphoid tissues and macrophages (
), suggesting that G2A plays a critical role in controlling peripheral lymphocyte homeostasis.
Controversial findings have been reported on endogenous ligands of G2A. At first, lysophosphatidylcholine and sphingosylphosphorylcholine were reported as potent ligands for G2A (
), along with three other related GPCRs, that is, OGR1, GPR4, and TDAG8. These receptors mediate accumulation of intracellular inositol phosphates or cAMP in response to acidic pH. However, G2A was reported to be less sensitive to pH fluctuations than the other three receptors in immune cells (
We recently reported another function of G2A as a receptor for oxidized free fatty acids, such as 9-hydroxyoctadecadienoic acid (HODE) and 11-hydroxyeicosatetraenoic acid (
). When G2A was expressed in CHO-K1 or HEK293 cells, 9-HODE induced [35S]GTPγS binding, intracellular calcium mobilization, inhibition of cAMP accumulation, and activation of a mitogen-activated protein kinase, JNK. However, the biological significance of G2A still remains unsolved.
While oxidized free fatty acids can be produced by many kinds of oxidative stresses, G2A was identified as a stress-inducible GPCR (
). Therefore, oxidative stresses might induce both ligand production and receptor expression to mediate appropriate cellular responses. As skin is routinely or pathologically exposed to many oxidative stresses, oxidized free fatty acid-G2A signaling might play biological roles in various conditions evoked by UV radiation, chemical agents, inflammation, microorganism infection, and other oxidative stresses in skin.
In skin, UV radiation can cause sunburns, skin cancer, skin aging, and immune suppression (
). Under the conditions with UVB radiation in skin, lipids containing linoleic acid might be oxidized, resulting in the production of 9-HODE, which acts as a ligand of G2A. In this study, we examined the effects of 9(S)-HODE on human keratinocytes, and the involvement of G2A in 9(S)-HODE-induced effects. We found that 9(S)-HODE evoked intracellular calcium mobilization and secretion of cytokines, which were enhanced by overexpression of G2A. 9(S)-HODE inhibited proliferation of keratinocytes due to cell cycle arrest in the G0/1-phase. Furthermore, UV and H2O2 caused G2A induction and 9-HODE production, and the UV-induced G2A was functional in secretion of cytokines. These results suggest that G2A plays biological roles as a receptor for 9-HODE under oxidative conditions in skin.
Results
Expression of G2A in human keratinocytes
First, the expression of G2A was examined in normal human skin by immunohistochemical analysis. Sections from shoulder skin were treated with a specific antibody against the second cytoplasmic domain of human G2A. G2A was expressed in the epidermis, preferentially in the spinous and granular cell layers, whereas its expression was lower in the basal cell layer (Figure 1a, upper sides). G2A was preferentially expressed on the cell membrane of normal human epidermal keratinocytes (NHEK) and an immortalized human keratinocyte cell line (HaCaT) (Figure 1a, middle and lower left sides). Competition studies of the G2A antibody with blocking peptide decreased the signals observed in both cells (Figure 1a, middle and lower right sides). G2A protein was detected as a single band of ∼42 kDa, compatible with the molecular weight of 42,368 calculated from the amino acid sequence of human G2A, in the membrane fractions of NHEK and HaCaT cells by western blotting, and the pretreatment with blocking peptide to the G2A antibody resulted in elimination of the band (Figure 1b). These data suggested that the antibody used specifically bound to G2A protein. Expression of G2A mRNA in cultured human keratinocytes was also examined. Total RNA was extracted from NHEK and HaCaT cells, and reverse transcriptase (RT)-PCR analysis was performed. G2A mRNA was expressed in both types of keratinocytes (Figure 1c). The level of G2A mRNA in NHEK cells at the fourth passage was estimated to be 104 order copies per μg of total RNA using quantitative real-time RT-PCR (data not shown).
Figure 1Expression of G2A in human keratinocytes. (a) Immunostaining for G2A in normal human skin, NHEK, and HaCaT cells. The primary antibody was a specific antibody against G2A or normal control rabbit IgG. The nuclei of NHEK and HaCaT cells were stained with SYTOX Orange Nucleic Acid Stain. For the competition study of G2A antibody, the primary antibody was preabsorbed with G2A blocking peptide. Bar=20 μm. (b) Western blotting of the membrane fractions (0.5 μg per lane) from NHEK and HaCaT cells to detect G2A or Na+/K+ ATPase. (c) Detection of G2A mRNA in cultured human keratinocytes. Expression of G2A mRNA was examined by PCR with or without RT reaction. The pCXN2.1-G2A vector was used as a template for positive controls. Data are representative of three independent experiments.
Intracellular calcium mobilization evoked by 9(S)-HODE
Next, we examined whether 9(S)-HODE could mediate any intracellular signals in NHEK cells, which endogenously express G2A. The intracellular calcium concentration was significantly increased with 10 μM 9(S)-HODE (Figure 2a). Sequential application of 15 μM 9(S)-HODE did not evoke any further responses, possibly due to receptor desensitization, as the cells could still respond to 10 μM ATP. In dose–response experiments, the calcium increase with 3 μM of 9(S)-HODE was significant and the maximal response was observed with 10 μM of 9(S)-HODE (Figure 2b).
Figure 2Intracellular calcium mobilization via G2A evoked by 9(S)-HODE. (a) Intracellular calcium mobilization in NHEK cells measured by an RF5300PC spectrofluorometer. Cells loaded with Fura-2/AM were stimulated with increasing concentrations of 9(S)-HODE. As a positive control, cells were stimulated with 10 μM ATP. (b) Calcium responses with various concentrations of 9(S)-HODE in NHEK cells (n=4, *P<0.01 (Student's t-test) vs corresponding values of 1 μM). Before stimulation, 9(S)-HODE was dissolved in HEPES-Tyrode's-BSA buffer after evaporation of ethanol under nitrogen gas. (c) Cell surface expression of exogenous G2A. Stable transformants of HaCaT cells that overexpress FLAG-tagged G2A were established (HaCaT-G2A). FLAG-tagged G2A was detected by flow cytometry using an M5 anti-FLAG antibody. Dotted line, HaCaT cells; solid line, HaCaT-G2A cells. (d) Intracellular calcium mobilization in HaCaT and HaCaT-G2A cells. Cells were stimulated with 10 μM 9(S)-HODE and 10 μM ATP. (e) Quantification of changes in intracellular calcium concentrations after stimulation with HODEs at the concentration of 10 μM. Data represent mean+SD (n=4, *P<0.01 (Student's t-test)). Data are representative of three independent experiments. Ex, excitation.
To confirm the involvement of G2A in 9(S)-HODE-induced calcium mobilization, we analyzed calcium responses in HaCaT cells. As parental HaCaT cells did not show any significant responses up to 10 μM 9(S)-HODE, a polyclonal population of HaCaT cells that stably overexpress FLAG-tagged G2A (HaCaT-G2A) was established. Cell surface expression of FLAG-tagged G2A proteins was confirmed by flow cytometric analysis using an anti-FLAG antibody (Figure 2c). Stable expression of G2A in HaCaT cells caused a significant calcium response with 9(S)-HODE (Figure 2d). 9(S)-HODE at 10 μM evoked an increase in the intracellular calcium concentration by 51.6±5.25 nM (Figure 2e, mean±SD; n=4). The response to 13(S)-HODE, less active ligand for G2A (
), was less than half of that to 9(S)-HODE in HaCaT-G2A cells (Figure 2e). These results indicate that 9(S)-HODE induced intracellular calcium mobilization through G2A in NHEK and HaCaT cells.
Cytokine secretion evoked by 9(S)-HODE in NHEK cells
Keratinocytes have been reported to produce various cytokines, including IL-1, -6, -8, -10, -12, GM-CSF, tumor necrosis factor-α, and so on, in response to diverse stimuli (
). We examined whether 9(S)-HODE induced cytokine secretion in NHEK cells. Among various cytokines examined, 9(S)-HODE induced productions of IL-6 (Figure 3a), IL-8 (Figure 3b), and GM-CSF (Figure 3c) in dose-dependent manners. Elevation of cytokine levels became evident at 4 hours in IL-6 and IL-8, and 16 hours in GM-CSF after 9(S)-HODE treatment. As shown in Figure 3d and e, 10 μM 13(S)-HODE did not induce productions of IL-6 and IL-8 8 hours after the treatment. These data indicate that 9(S)-HODE induces cytokine secretion in NHEK cells via G2A but not through such mechanisms as toxic effects of oxidized free fatty acids.
Figure 3Cytokine secretion evoked by 9(S)-HODE in NHEK cells. (a–c) Cytokine productions evoked by various concentrations of 9(S)-HODE. (a) IL-6; (b) IL-8; (c) GM-CSF (n=3, *P<0.05 (Student's t-test) vs corresponding values of vehicle control). (d, e) Cytokine productions evoked by HODEs at the concentration of 10 μM 8 hours after the treatment. (d) IL-6; (e) IL-8 (*P<0.01, **P<0.05 (Student's t-test)). Data are representative of three independent experiments.
Involvement of G2A in cytokine secretion evoked by 9(S)-HODE in HaCaT cells
To examine the involvement of G2A in 9(S)-HODE-induced cytokine secretion, we analyzed 9(S)-HODE-induced IL-6 and IL-8 secretion in HaCaT-G2A cells. While HODEs did not induce cytokine secretion in parental HaCaT cells, overexpression of G2A in HaCaT cells resulted in significant increases in IL-6 and IL-8 releases 8 hours after 9(S)-HODE treatment (Figure S1, right sides, filled squares). The amounts of both cytokines induced by 13(S)-HODE were much smaller than those induced by 9(S)-HODE in HaCaT-G2A cells (Figure S1, right sides, striped squares). Without HODEs treatment, both cytokine levels in the culture supernatant of HaCaT-G2A cells were higher than those of HaCaT cells (Figure S1, open squares), possibly due to the constitutive activity of G2A as shown in HeLa cells exogenously expressing G2A (
). Taken together with the results in NHEK cells (Figure 3), these results indicate that cytokine secretion evoked by 9(S)-HODE was mediated via G2A in keratinocytes.
Involvement of G2A in cytokine secretion evoked by 9(S)-HODE in HaCaT cells. HaCaT and HaCaT-G2A cells were serum-starved for 4 h, and cultured with HODEs at the concentration of 10 μM in DMEM for 8 h. The culture supernatants were collected, and concentrations of cytokines were measured using a Bio-Plex ELISA system. (a) IL-6; (b) IL-8. Data represent mean + SD (n = 3, * p < 0.01, ** p < 0.05 (Student's t-test)), and are representative of three independent experiments.
Inhibition of proliferation and morphological changes by 9(S)-HODE in NHEK cells
Next, we examined whether 9(S)-HODE had any effects on the proliferation of keratinocytes. Cell viability was determined by measuring the contents of ATP in the cells. Cell viability was decreased by 9(S)-HODE treatment in a dose-dependent manner (Figure 4a). The treatment with 10 μM 9(S)-HODE strongly decreased cell viability, and this effect continued for 48 hours accompanied by morphological changes in NHEK cells (Figure 4b). The light microscopic images, taken 24 hours after 9(S)-HODE treatment, showed that 9(S)-HODE-treated cells contained enlarged cytoplasm and distinct shiny perinuclear vesicles with a diameter of approximately 1 μm. The cytoplasm swelling was more apparent 48 hours after 9(S)-HODE treatment, and elongated cells appeared with the shiny granules located on top of keratinocyte colonies. On the other hand, the treatment with 10 μM 13(S)-HODE did not affect the cell viability (Figure 4c) and neither did it change the morphology of NHEK cells (data not shown). To examine the involvement of G2A in 9(S)-HODE-induced decrease of cell viability in NHEK cells, we tried to suppress the expression of G2A using small interfering RNA (siRNA). An siRNA, named siRNA-132, decreased the expression of G2A mRNA to 61% of that of scrambled control cells in a quantitative real-time RT-PCR analysis (Figure 4d). By this siRNA-mediated suppression of G2A, the decrease of the cell viability 24 hours after 9(S)-HODE treatment was partially inhibited (Figure 4e). These data indicate that 9(S)-HODE induced suppression of proliferation via G2A in NHEK cells. On the other hand, cell viability was not significantly affected in both HaCaT and HaCaT-G2A cells by the treatment with 10 μM 9(S)-HODE (discussed later).
Figure 4Inhibition of proliferation by 9(S)-HODE in NHEK cells. (a) Cell viability of NHEK cells treated with various concentrations of 9(S)-HODE in the growth medium (n=4, *P<0.01 (Student's t-test) vs corresponding values of vehicle control). RLU, relative luminescence units. (b) Morphology of cells treated with 10 μM 9(S)-HODE. The open arrowheads show typical shiny perinuclear vesicles. Original magnification is 400-fold. Bar=20 μm. (c) Cell viability of NHEK cells treated with HODEs at the concentration of 10 μM (n=4, *P<0.01 (Student's t-test) vs corresponding values of vehicle control). (d) Effects of siRNA-132 on the expression of G2A mRNA in NHEK cells 48 hours after transfection of siRNAs (n=3, *P<0.01 (Student's t-test) vs corresponding values of scrambled control). (e) siRNA-mediated inhibition of 9(S)-HODE-induced effects on cell viability of NHEK cells 24 hours after 10 μM 9(S)-HODE treatment (n=3, *P<0.01 (Student's t-test)). Data are representative of three independent experiments.
Suppression of DNA synthesis and cell cycle arrest in the G0/1-phase by 9(S)-HODE in NHEK cells
To elucidate the mechanisms of 9(S)-HODE-induced inhibition of the proliferation in NHEK cells, cell cycle analysis was performed. NHEK cells were treated with 10 μM 9(S)-HODE for 24 hours and stained with propidium iodide, followed by flow cytometric analysis of the cell cycle distribution. 9(S)-HODE increased the cell percentage in the G0/1-phase by 20% and decreased the cell percentage in the S-phase by more than 10% (Figure S2a and b). The cell percentage of sub-G1 cells, apoptotic cells with degraded DNA, was not changed by 9(S)-HODE treatment, while it increased after 10 mJ cm−2 UVB irradiation. DNA fragmentation was not detected in cells treated with 10 μM 9(S)-HODE, while it was obvious in the cells irradiated with UVB (Figure S2c). To examine the effects of 9(S)-HODE on DNA synthesis in NHEK cells, incorporation of BrdU, an analogue of thymidine, was measured. Treatment with 10 μM 9(S)-HODE decreased BrdU incorporation to 40% of that of vehicle control in NHEK cells (Figure S2d). These results suggested that 9(S)-HODE inhibited proliferation of NHEK cells, which was caused by suppression of DNA synthesis and cell cycle arrest in the G0/1-phase, but not by apoptosis.
Suppression of DNA synthesis and cell cycle arrest in the G0/1-phase by 9(S)-HODE in NHEK cells. (a) Cell cycle analysis of propidium iodide-stained cells by flow cytometry 24 h after each stimulation. Data represent percentages of total cells. (b) Statistical analysis of (a). Open squares, vehicle control; filled squares, 10 μM 9(S)-HODE; striped squares, 10 mJ/cm2 UVB. Data represent mean + SD (n = 4). * p < 0.01 (Student's t-test) vs. corresponding values of vehicle control. (c) DNA fragmentation assay 24 h after the stimulation of 10 μM 9(S)-HODE and 10 mJ/cm2 UVB. (d) BrdU incorporation assay. NHEK cells were cultured in the basal medium for 2 h, and treated with 10 μM 9(S)-HODE for 18 h in the growth medium. Amounts of incorporated BrdU were determined. Data represent mean + SD (n = 4). * p < 0.01 (Student's t-test) vs. corresponding values of vehicle control. Data are representative of three independent experiments.
Induction of G2A by UVB and H2O2
Since G2A was reported as a DNA damage- and stress-inducible GPCR (
), we examined levels of G2A mRNA in cultured keratinocytes irradiated with various doses of UVB. UVB increased G2A mRNA in a dose-dependent manner in HaCaT cells 24 hours after irradiation up to 10 mJ cm−2. The level of G2A mRNA in 10 mJ cm−2 UVB-irradiated cells was nearly double than that in mock-irradiated cells (Figure 5a). Hydrogen peroxide (H2O2) treatment at 100 μM also increased G2A mRNA in HaCaT cells (Figure 5b). The levels of glyceraldehyde-3-phosphate dehydrogenase mRNA were simultaneously analyzed, but they did not change in HaCaT cells under these experimental conditions (data not shown). Next, ROS productions were analyzed using CM-H2DCFDA dye, which was oxidized by ROS to the highly fluorescent 2′,7′-dichlorofluorescein. The fluorescence intensity was increased by both UVB and H2O2 in HaCaT cells (Figure 5c). The increase became significant immediately after UVB irradiation and 30 minutes after H2O2 treatment, respectively. Thus, the induction of G2A by these two stimuli was accompanied by ROS production.
Figure 5Induction of G2A by UVB and H2O2. (a, b) Expression levels of G2A mRNA in HaCaT cells 24 hours after (a) UVB irradiation or (b) H2O2 treatment. A quantitative real-time RT-PCR analysis was performed (n=4, *P<0.01 (Student's t-test) vs corresponding values of (a) mock-irradiated cells, and (b) unstimulated control). (c) Detection of ROS production after UVB irradiation or H2O2 treatment in HaCaT cells (n=4, *P<0.01, **P<0.05 (Student's t-test) vs corresponding values of unstimulated control). RFU, relative fluorescence units. (d) Increase of 9(S)-HODE-induced IL-6 production after UVB irradiation in HaCaT cells. HaCaT cells were irradiated (10 mJ cm−2 UVB) or mock-irradiated in PBS and cultured in DMEM containing 10% fetal bovine serum. After 12 hours, cells were serum-starved and cultured for another 12 hours. The cells were then treated with 10 μM 9(S)-HODE in DMEM, and concentrations of IL-6 at 8 hours were measured using a Bio-Plex ELISA system (n=3, *P<0.05 (Student's t-test)). Data are representative of three independent experiments.
Next, we examined whether UVB-induced G2A enhanced cytokine secretion evoked by 9(S)-HODE. HaCaT cells were irradiated with 10 mJ cm−2 UVB, incubated for 24 hours, and treated with 10 μM 9(S)-HODE. As shown in Figure 5d, the level of IL-6 8 hours after 9(S)-HODE treatment was significantly higher in UVB-irradiated cells than in mock-irradiated cells (see filled squares). Without 9(S)-HODE treatment, IL-6 production was increased in UVB-irradiated cells (see open squares) probably due to the constitutive activity of the induced G2A as shown in Figure S1. These data indicate that functional G2A was induced by UVB irradiation in keratinocytes.
Production of 9-HODE by UVB and H2O2
Finally, we examined whether oxidative stresses cause conversion of linoleic acid to 9-HODE. Among the derivatives of linoleic acid, 9-HODE and 9-hydroperoxyoctadecadienoic acid were the most potent ligands of G2A, while 13-HODE and 13-hydroperoxyoctadecadienoic acid were much less active (
). We quantified the amounts of HODEs (9-(E,Z)-HODE and 13-(Z,E)-HODE) in UVB-irradiated linoleic acid by liquid chromatography-mass spectrometry using deuterated-9-HODE as an internal standard after chemical reduction of hydroperoxyoctadecadienoic acids to HODEs (Supplementary Materials and Methods). As shown in Figure S3, 100 mJ cm−2 UVB increased the level of 9-HODE. At the same time, 13-HODE was produced at comparable level to 9-HODE. The increase became prominent 2 minutes after the irradiation (9-HODE, about 4.1 ng per dish; Figure S3a) and reached a plateau after 10 minutes (9-HODE, about 6.0 ng per dish; Figure S3a). The production of HODEs was dependent on the dose of UVB (Figure S3b). The dose required for the significant increase of 9-HODE was 25 mJ cm−2, but 10 mJ cm−2 UVB had a tendency to produce 9-HODE. These results suggest that G2A ligands are produced from polyunsaturated fatty acids, such as linoleic acid, possibly through radical reaction, when they are directly exposed to UVB. Furthermore, we examined whether UVB caused 9-HODE production in HaCaT cells. As cultured keratinocytes are thought to be in an essential fatty acid-deficient state (
), we supplemented HaCaT cells with linoleic acid before UVB irradiation to observe linoleate oxidation sensitively. Total lipids extracted from stimulated cells were analyzed by liquid chromatography-mass spectrometry after treatment with phospholipase A2 and chemical reduction. As shown in Figure S3c, the levels of 9-HODE and 13-HODE were increased approximately twofold 30 minutes after UVB irradiation (100 mJ cm−2). Treatment with 200 μM H2O2 also resulted in increased oxidation of linoleate (approximately threefold; Figure S3c). The increase was not observed without phospholipase A2 treatment (data not shown), suggesting that linoleate was oxidized in its esterified form in phospholipids. These data indicate that keratinocytes might produce G2A ligands under oxidative stresses.
Production of 9-HODE by UVB and H2O2. (a)(b) The conversion of free linoleic acid to 9-HODE by UVB irradiation. Linoleic acid in a 60 mm dish was irradiated with 100 mJ/cm2 UVB, and incubated for 0-30 min (a), or irradiated with increasing doses of UVB, and incubated for 10 min (b). (n = 3, * p < 0.01, ** p < 0.05 (Student's t-test) vs. corresponding values at 0 min (a), and of mock-irradiated samples (b), respectively). (c) Oxidation of linoleate by UVB or H2O2 in HaCaT cells loaded with linoleic acid. HaCaT cells in a 100 mm dish were stimulated with 100 mJ/cm2 UVB or 200 μM H2O2, and incubated for 30 min at room temperature. Total lipids were extracted and digested with phospholipase A2, followed by reduction with NaBH4. Contents of HODEs were analyzed using a liquid chromatography-mass spectrometry system. (n = 4, * p < 0.01 (Student's t-test) vs. corresponding values of unstimulated control). Data are representative of three independent experiments.
Discussion
G2A was first identified as a GPCR that could be induced by various DNA-damaging and stress-inducing stimuli, including UV irradiation in human Ramos B cells (
). In this study, we showed that G2A was also expressed in human epidermis and keratinocytes (Figure 1), and the expression was increased by UVB and H2O2 (Figure 5). We further showed that these oxidative stimuli caused 9-HODE production (Figure S3). 9(S)-HODE induced intracellular calcium mobilization (Figure 2), cytokine production (Figure 3), and inhibition of proliferation (Figure 4) in keratinocytes, while 13(S)-HODE revealed no effect or much smaller effects than 9(S)-HODE (Figures 2e, 3d, e, and 4c, and Figure S1). These responses were enhanced by overexpression or induction of G2A (Figures 2d, e, and 5d, and Figure S1) and were attenuated by decrease of G2A (Figure 4e). Thus, we conclude that these 9(S)-HODE-induced responses are mediated via G2A and assume that G2A plays biological roles in skin as a receptor for 9-HODE.
There were some differences in response to 9-HODE between NHEK and HaCaT cells. Although HaCaT cells seemed to possess more G2A protein than NHEK cells (Figure 1b), HaCaT cells showed much smaller responses to 9-HODE than NHEK cells in cytokine production (Figure S1) and no response in calcium mobilization (Figure 2d). HaCaT cells came to show significant responses to 9-HODE after expressing exogenous G2A in HaCaT-G2A cells (Figure 2e and Figure S1). The reason why parent HaCaT cells did not show apparent response to 9-HODE remains to be elucidated.
Some oxidative derivatives of linoleic and arachidonic acids can activate G2A. Among them, free 9-HODE is the most potent in mediating intracellular calcium mobilization in CHO-K1 cells expressing G2A (
). 9-HODE can be produced by oxidation of linoleic acid through both radical and enzymatic reactions, and linoleic acid is the most commonly found polyunsaturated fatty acid in epidermis (
). Oxidative stresses, such as UV radiation, gaseous pollutants, chemical agents, microorganism infection, skin cancer, and inflammations, are supposed to induce radical reactions in skin and lead to induction of cutaneous lipid peroxidation with concomitant modulation in the level of antioxidant and drug-metabolizing enzymes (
). For instance, free radical oxidation was reported as the main process in producing oxidized derivatives of linoleic acid, including 9-HODE in mouse skin treated with phorbol myristate acetate (
). Furthermore, enzymatic reaction of prostaglandin H-synthase-2 has been found to convert linoleic acid to 9-HODE in mouse keratinocytes and human dermal fibroblasts under stimulation of epidermal growth factor (
Prostaglandin H-synthase-2 is the main enzyme involved in the biosynthesis of octadecanoids from linoleic acid in human dermal fibroblasts stimulated with interleukin-1beta.
), respectively. Some of the lipoxygenases and cytochrome P450 enzymes would also be involved in the metabolism of linoleic and arachidonic acids in skin.
Linoleic acid is known to be essential for proper cutaneous barrier function, and linoleic acid-rich acylceramide is believed to have a specific function in the formation of stratum corneum lipid layers (
). In addition to ceramides, cholesterol and free fatty acids are rich in the stratum corneum, and linoleic acid is the most abundant polyunsaturated fatty acid in free fatty acid fractions (
). In this context, when skin is exposed to oxidative stresses, free linoleic acid might be converted to free 9-HODE in the uppermost layer of the epidermis as shown in Figure S3a and b. Also, linoleic acid-containing acylceramide and cholesterol ester in the stratum corneum could be oxidized and then hydrolyzed to produce free 9-HODE.
In epidermal cell layers, linoleic acid exists abundantly as linoleate esterified to phospholipids of cell membranes in the suprabasal and basal cell layers (
). In this study, the oxidation of linoleate in the membrane lipid of HaCaT cells by UVB and H2O2 was observed after the in vitro treatment with phospholipase A2 (Figure S3c). In skin under oxidative conditions, oxidized linoleate could be hydrolyzed to produce HODEs by phospholipase A2 that might be activated by ROS signaling pathway. In psoriatic skin, 9-HODE has been detected as a free acid (
), and levels of 9-HODE esterified to the sn-2 position of phospholipids in lesional psoriatic skin are significantly decreased compared with non-lesional psoriatic skin (
). It is possible that membrane phospholipids are hydrolyzed by certain types of phospholipase A2 generating free 9-HODE in lesional psoriatic skin. An epithelium-specific cytosolic phospholipase A2 (cytosolic PLA2δ) was reported to be induced in psoriatic skin (
Many of the UVB-induced damages in keratinocytes might be mediated by ROS and subsequently induced cytokines. Production of ROS causes oxidation of lipids. In this report, UVB was found to cause both production of 9-HODE (Figure S3) and induction of G2A expression (Figure 5a) accompanied by ROS production (Figure 5c) in keratinocytes. Cytokine secretion evoked by 9(S)-HODE was enhanced in cells irradiated with UVB (Figure 5d). Intensity of UVB (5 mJ cm−2) required for G2A induction is lower than those used in previous reports in which effects of UVB irradiation on keratinocytes were analyzed (
). Although higher doses of UVB (25 mJ cm−2) were required for the significant increase of 9-HODE converted from linoleic acid, the doses are still in the range we are commonly exposed to in daily life (
). In case of sunburn, UV irradiation first occurs in stratum corneum that contains linoleic acid-rich lipid layers as discussed above. Then, diminished UV may still have some biological effects in keratinocytes in the granular and spinous cell layers where G2A mainly exists (Figure 1a).
Keratinocytes participate in immune systems by producing various cytokines in response to diverse stimuli. Cytokines released from keratinocytes affect Langerhans cells, lymphocytes, vascular endothelial cells, and keratinocytes themselves to regulate immune responses and inflammatory reactions (
). In this study, productions of IL-6 and IL-8 were observed in keratinocytes treated with 9(S)-HODE (Figure 3a and b), and the levels of IL-6 and IL-8 were increased by overexpression of G2A (Figure S1). IL-6 induces keratinocyte proliferation in vitro (
). On the other hand, IL-8 is a chemokine, which recruits neutrophils, macrophages, and T cells, whereas IL-8 also promotes keratinocyte proliferation (
) would be associated with the hyperproliferation of keratinocytes and accumulation of neutrophils, which are characteristic findings of psoriatic lesions. In addition to IL-6 and IL-8, GM-CSF was released from keratinocytes by 9(S)-HODE stimulation (Figure 3c). 9-HODE was reported to be a strong proinflammatory mediator in an experimental wound-healing model of the rat (
The linoleic acid metabolite 9DS-hydroxy-10,12(E,Z)-octadecadienoic acid is a strong proinflammatory mediator in an experimental wound healing model of the rat.
). G2A might be involved in the initiation or progression of various pathological conditions, including psoriasis, lichen planus, and pustulosis through cytokine secretion from keratinocytes, affecting keratinocytes and inflammatory cells.
9(S)-HODE inhibited proliferation of keratinocytes, which was caused by suppression of DNA synthesis and cell cycle arrest in the G0/1-phase (Figure 4 and Figure S2). G2A overexpression was reported to cause cell accumulation at the G2/M-phase in NIH3T3 cells, although the possibility that G2A might be functional at other checkpoints in the cell cycle was not excluded (
). It is reasonable to assume that cell cycle arrest is induced under cell-damaging conditions such as oxidative stresses. In this study, 9(S)-HODE did not inhibit proliferation of HaCaT cells (data not shown). We suppose that HaCaT cells might not be suitable to examine the effects of 9-HODE on cell proliferation because the p53 mutations, which largely affect the signals of cell cycle, were identified in HaCaT cells (
), it was inferred that some of the effects induced by 9(S)-HODE in NHEK cells were mediated via PPARγ. Involvement of PPARγ has been implicated in the regulation of inflammatory processes and cytokine release in various cell types. To examine this possibility, NHEK cells were treated with PPARγ antagonist, GW9662, and assessed using a cell viability assay and a cytokine assay. GW9662 (10 μM) did not inhibit the effects induced by 10 μM 9(S)-HODE (data not shown). In a previous study, the required concentration of 9(S)-HODE to induce reporter activities of PPARγ was higher than 10 μM in macrophages (
). In contrast, 1 μM 9(S)-HODE was effective for production of cytokines (Figure 3) and for inhibition of cell proliferation (Figure 4a) in keratinocytes. Furthermore, the expression of PPARγ is low in undifferentiated keratinocytes, although it is increased during differentiation (
). Taken together, we conclude that the effects of 9(S)-HODE observed in this study were mediated via G2A, although it is still possible that PPARγ might have additive, synergistic, or modifying effects during differentiation.
In summary, we showed that G2A was expressed in keratinocytes and was induced by UVB and H2O2. These oxidative stimuli also caused oxidation of linoleate. G2A was involved in 9(S)-HODE-induced intracellular calcium mobilization, secretion of inflammatory cytokines, and inhibition of cell proliferation in keratinocytes. G2A may play proinflammatory roles as a receptor for oxidized free fatty acids such as 9-HODE under oxidative pathological conditions in skin.
Materials and Methods
Materials
Linoleic acid was purchased from Sigma (St Louis, MO). 9(S)-HODE was synthesized from linoleic acid by an enzymatic reaction with potato 5-lipoxygenase as described previously (
). 13(S)-HODE was purchased from Cayman Chemical (Ann Arbor, MI). Ethanol (0.1%) was used as a vehicle for HODEs. An expression vector (pCXN2.1-G2A) for FLAG-tagged human G2A (NCBI accession number AF083955) was constructed as described previously (
Normal human skin tissues were obtained from surgically excised skin with written informed consent. The study was performed in accordance with institutional guidelines set forth by Gunma University School of Medicine and adhered to the Declaration of Helsinki Principles for use of human tissue. Frozen sections (6-μm thick) were blocked in 10% BSA at room temperature for 30 minutes and then incubated with primary antibodies at 4°C overnight. Primary antibodies (1.7 μg ml−1) were either rabbit anti-human G2A antibody (MBL, Woburn, MA) or normal rabbit IgG (Santa Cruz Biotechnology, San Diego, CA) for the negative control sections. Following incubation with 2 μg ml−1 secondary antibody (goat anti-rabbit IgG conjugated with Alexa Fluor 488; Molecular Probes, Eugene, OR) at room temperature for 1 hour, fluorescence images were collected using a microscope (Axioskop; Carl Zeiss, Oberkochen, Germany). In case of cell staining, suspended NHEK and HaCaT cells were attached to the slide glasses by centrifugation (300 r.p.m., 2 minutes), fixed with 100% methanol, permeabilized with 0.1% Triton X100 for 30 minutes, and blocked in 1% BSA. The cells were then incubated with primary antibody (1.7 μg ml−1) at room temperature for 30 minutes. For the competition study of the G2A antibody, cells were stained with the primary antibody preincubated with 10-fold molar volume of G2A blocking peptide (MBL) at room temperature for 60 minutes. Following incubation with 0.67 μg ml−1 secondary antibody and 1 μM SYTOX Orange Nucleic Acid Stain (Molecular Probes) at room temperature for 30 minutes, fluorescence images were collected using a confocal microscope (LSM510; Carl Zeiss). Antibodies were diluted in phosphate-buffered saline (PBS) containing 1% BSA.
Western blotting
NHEK cells were disrupted by sonication in a homogenizing buffer (20 mM Tris-HCl, pH 7.4, 0.25 M sucrose, 10 mM MgCl2, 2 mM EDTA, and Complete protease inhibitor mixture (Roche Applied Science, Indianapolis, IN)). The homogenates were centrifuged for 10 minutes at 800 × g, and the resulting supernatants were further centrifuged for 60 minutes at 100,000 × g. The precipitates (membrane fractions) were dissolved in the homogenizing buffer containing 0.5% dodecylmaltoside (Dojindo, Kumamoto, Japan). An aliquot of protein (0.5 μg per lane) was separated on 10% SDS-polyacrylamide gel electrophoresis, transferred to a Hybond-P polyvinylidene difluoride membrane (GE Healthcare, Buckinghamshire, UK), blocked in Tris-buffered saline containing 50% Block Ace (Dainippon Sumitomo Pharma, Osaka, Japan) at 4 °C overnight, and then incubated with the rabbit anti-human G2A antibody (0.5 μg ml−1) or a 500-fold diluted mouse ascites M7-PB-E9, including anti-Na+/K+-ATPase monoclonal antibody (Sigma) for 1 hour at room temperature. For the competition study of the G2A antibody, the primary antibody was preincubated with 100-fold molar volume of G2A blocking peptide. After incubation with each horseradish peroxidase-conjugated secondary antibody (80 ng ml−1) (Santa Cruz Biotechnology), the signals were visualized using an ECL plus Western blotting Detection System (GE Healthcare).
Cell culture
NHEK from newborn foreskin were obtained from Kurabo (Osaka, Japan) and were maintained in HuMedia-KB2 (Kurabo, basal medium), supplemented with 10 μg ml−1 insulin, 0.1 ng ml−1 human epidermal growth factor, 0.5 μg ml−1 hydrocortisone, 50 μg ml−1 gentamicin, 50 ng ml−1 amphotericin B, and 0.4% bovine pituitary extract (growth medium). Usually, NHEK cells between third and fifth passages were used for experiments. Immortalized human keratinocyte cell line HaCaT (
) was maintained in DMEM (Sigma) containing 10% fetal bovine serum. Cells were cultured at 37 °C in a humidified incubator with 5% CO2.
Transfection and stable expression of G2A
HaCaT cells were transfected with pCXN2.1-G2A using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Stably transfected clones resistant to 1 mg ml−1 Geneticin (Invitrogen) were collected, and expression levels of G2A were confirmed by RT-PCR and flow cytometric analyses. To observe expression of FLAG-tagged G2A proteins on cell surfaces, cells were incubated with 10 μg ml−1 M5 anti-FLAG antibody (Sigma) without cell permeabilization, followed by staining with 10 μg ml−1 goat anti-mouse IgG conjugated with Alexa Fluor 488 (Molecular Probes), and were analyzed using an EPICS XL flow cytometer system (Beckman Coulter, Fullerton, CA). Cells were treated with each antibody in PBS containing 1% BSA at room temperature for 1 hour.
RT-PCR analysis
Total RNA was extracted from NHEK or HaCaT cells using an RNeasy Mini kit (Qiagen, Hilden, Germany) with on-column DNase digestion. RT-PCR analysis of G2A mRNA was performed using a QIAquick one step RT-PCR kit (Qiagen) with the following primer set: sense primer, 5′-GGCTTTGCCATCCCTCTC-3′; antisense primer, 5′-GACAGGCACAGAAACACC-3′. Quantitative real-time RT-PCR analysis was performed using a DNA Engine Opticon 2 system (MJ Research, Waltham, MA).
Measurement of intracellular calcium concentration
NHEK cells were loaded with 2.5 μMFura-2/AM (Dojindo) in Hepes-Tyrode's-BSA buffer (25 mM Hepes-NaOH, pH 7.4, 140 mM NaCl, 2.7 mM KCl, 1.0 mM CaCl2, 12 mM NaHCO3, 5.6 mM D-glucose, 0.37 mM NaH2PO4, 0.49 mM MgCl2, and 0.01% fatty acid-free BSA) containing 0.02% pluronic F127 for 1 hour at 37°C. In case of HaCaT cells, the concentrations of Fura-2/AM and F127 were 5 μM and 0.04%, respectively. Cells were washed with Hepes-Tyrode's-BSA buffer, and changes in intracellular calcium concentrations upon ligand stimulation were monitored with an RF5300PC spectrofluorometer (Shimadu, Kyoto, Japan).
Determination of cytokine concentrations
NHEK and HaCaT cells were treated with HODEs in the basal medium and DMEM, respectively. HaCaT cells were serum-starved before 9(S)-HODE treatment. The culture supernatants were collected at indicated times after HODEs treatment, and concentrations of cytokines were measured using a Bio-Plex ELISA system (BioRad, Hercules, CA).
Viability and morphology of NHEK cells
NHEK cells were treated with various concentrations of HODEs in the growth medium. Cell viability was determined by luminescent signals proportional to amount of ATP in the cells at indicated times using a CellTiter-Glo Luminescent Cell Viability Assay kit (Promega, Madison, WI), according to the manufacturer's instructions. Cell morphology was observed using a microscope (CKX41; Olympus, Tokyo, Japan).
Suppression of G2A by siRNA
NHEK cells were transfected with siRNAs (Samchully Pharm, Seoul, Korea) using Lipofectamine 2000 reagent. The sequences of G2A-specific (named siRNA-132) and scrambled control siRNAs were 5′-GGUACUACUACGCCAGGUUTTAACCUGGCGUAGUAGUACCTT-3′, and 5′-GCUUCUAGUACGCGAGGAUTTAUCCUCGCGUACUAGAAGCTT-3′, respectively. After 24 hours, NHEK cells were treated with 10 μM 9(S)-HODE for 24 hours in the growth medium, and cell viability was determined using a CellTiter-Glo Luminescent Cell Viability Assay kit. To confirm the effects of siRNA, the levels of G2A mRNA 48 hours after transfection of siRNA were examined by a quantitative real-time RT-PCR analysis using the following primer set: sense primer, 5′-CATCCTCGTCGGGATCGTTC-3′; antisense primer, 5′-GAGAGAGGGATGGCAAAGCC-3′.
Cell cycle analysis
NHEK cells were treated with 10 μM 9(S)-HODE or irradiated with 10 mJ cm−2 UVB and then cultured in the growth medium. After 24 hours, cells were fixed with 70% ethanol and incubated with 250 μg ml−1 RNase (Marligen Biosciences, Ijamsville, MD) in PBS containing 0.1% BSA at 37°C for 30 minutes. After staining with 50 μg ml−1 propidium iodide (Sigma) on ice for 30 minutes, cell cycles were analyzed using an EPICS XL flow cytometer system.
DNA fragmentation assay
NHEK cells were treated with 10 μM 9(S)-HODE or irradiated with 10 mJ cm−2 UVB and then cultured in the growth medium. After 24 hours, cells were resuspended in lysis buffer (10 mM Tris-HCl, pH 7.4, 10 mM EDTA, and 0.5% Triton X100) and centrifuged for 5 minutes at 15,000 r.p.m. The resulting supernatants were collected and then incubated with 200 μg ml−1 RNase at 37°C for 1 hour and 200 μg ml−1 proteinase K at 50°C for 30 minutes. DNA was extracted with isopropanol, followed by agarose gel electrophoresis analysis.
BrdU incorporation assay
NHEK cells were cultured in the basal medium for 2 hours and treated with 10 μM 9(S)-HODE in the growth medium for 18 hours. Amounts of incorporated BrdU were determined using a BrdU Labeling and Detection Kit III (Roche Applied Science) according to the manufacturer's instructions.
UVB irradiation
Cells in PBS were irradiated with UVB using two FL20S/E lamps (Toshiba, Tokyo, Japan) that emitted most of their energy within the UVB range, with an emission peak at 306 nm. The strength of UVB rays was determined with a Model J-260 digital radiometer (Ultra-Violet Products, Cambridge, UK). Mock-irradiated cells were treated in an identical manner, except that they were shaded from the UVB lamps.
Detection of ROS
HaCaT cells were loaded with 5 μM CM-H2DCFDA (Molecular Probes) in PBS at 37°C for 20 minutes. After washing with PBS, cells were stimulated with increasing doses of UVB or 100 μM H2O2. Intracellular production of ROS was monitored at indicated times with a FLEX station scanning fluorometer system (Molecular Devices, Sunnyvale, CA) at an excitation wavelength of 485 nm and an emission wavelength of 525 nm.
Conflict of Interest
The authors state no conflict of interest.
ACKNOWLEDGMENTS
We thank Ms Takeuchi Y. (Gunma University) for her expertise in immunohistochemistry, Dr Miyazaki J. (Osaka University) for providing the pCXN2.1 vector, and Drs Fusenig N.E. (German Cancer Research Center), Kuroki T. (University of Showa), and Nozawa Y. (Gifu International Institute of Biotechnology) for providing the HaCaT cells. This work was supported by Grants-in-Aid and the 21st Century Center of Excellence Program of the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
SUPPLEMENTARY MATERIAL
Materials and Methods. Detection of HODEs.
Figure S1. Involvement of G2A in cytokine secretion evoked by 9(S)-HODE in HaCaT cells.
Figure S2. Suppression of DNA synthesis and cell cycle arrest in the G0/1-phase by 9(S)-HODE in NHEK cells.
Figure S3. Production of 9-HODE by UVB and H2O2.
REFERENCES
Baer A.N.
Costello P.B.
Green F.A.
Free and esterified 13(R,S)-hydroxyoctadecadienoic acids: principal oxygenase products in psoriatic skin scales.
Prostaglandin H-synthase-2 is the main enzyme involved in the biosynthesis of octadecanoids from linoleic acid in human dermal fibroblasts stimulated with interleukin-1beta.
The linoleic acid metabolite 9DS-hydroxy-10,12(E,Z)-octadecadienoic acid is a strong proinflammatory mediator in an experimental wound healing model of the rat.
33854-9&id=gr1.jpg){kind=link}
33854-9&id=gr2.gif){kind=link}
33854-9&id=gr3.gif){kind=link}
33854-9&id=gr4.jpg){kind=link}
33854-9&id=gr5.gif){kind=link}