Prodrugs for Skin Delivery of Menahydroquinone-4, an Active Form of Vitamin K2(20), Could Overcome the Photoinstability and Phototoxicity of Vitamin K2(20)

The effective delivery of menahydroquinone-4 (MKH), an active form of menaquinone-4 (MK-4, vitamin K2(20)), to the skin is beneficial in the treatment of various skin pathologies. However, its delivery through the application of MK-4 to the skin is hampered due to the photoinstability and phototoxicity of MK-4. This study aimed to evaluate the potential of ester prodrugs of MKH for its delivery into the skin to avoid the abovementioned issues. The ester prodrugs, MKH 1,4-bis-N,N-dimethylglycinate hydrochloride (MKH-DMG) and MKH 1,4-bis-hemisuccinate (MKH-SUC), were prepared using our previously reported methods. Photostability was determined under artificial sunlight and multi-wavelength light irradiation, phototoxicity was determined by intracellular ROS formation and cell viability of UVA-irradiated human epidermal keratinocyte cells (HaCaT), and delivery of MKH into HaCaT cells was assessed by measuring menaquinone-4 epoxide (MKO) levels. MKH prodrugs showed higher photostability than MK-4. Although MK-4 induced cellular ROS and reduced cell viability after UVA irradiation, MKH prodrugs did not affect either ROS generation or cell viability. MKH prodrugs enhanced intracellular MKO, indicating effective delivery of MKH and subsequent carboxylation activity. In conclusion, these MKH prodrugs show potential for the delivery of MKH into the skin without photoinstability and phototoxicity.


Introduction
Menaquinone-4 (MK-4) or vitamin K 2 (20) is an important vitamin K compound used clinically. Menahydroquinone-4 (MKH), the fully reduced and active form of MK-4, is a cofactor for γ-glutamyl carboxylase (GGCX), which catalyzes the post-transcriptional carboxylation of vitamin K-dependent proteins. MKH is produced by the enzymatic conversion of vitamin K 1 and K 3 in the body [1]. Since MKH is readily oxidized to MK-4, MK-4 is available for clinical use.
Because vitamin K is highly unstable in light, it should be strictly protected from light exposure during manufacturing, preparation, storage, and treatment [10][11][12]. Furthermore, vitamin K is phototoxic against cultured epidermis upon ultraviolet A (UVA) irradiation (dose 6 J/cm 2 ) [13]. Given that when vitamin K is applied to exposed skin, it is difficult to protect it from light exposure, and these physicochemical properties of MK-4 limit its use as a skin application and hamper its beneficial effects.
It has been shown that the main photodegradation product of MK-4 is MK-4 chromenol (MKC), which is produced from the naphthoquinone structure and isoprenyl side chain of MK-4. We have previously reported that ester-type prodrugs of MKH, such as the 1,4-bis-N,N-dimethylglycinate (MKH-DMG) and hemi-succinate (MKH-SUC) derivatives, can act as delivery systems for MKH without a reductive activation step [14][15][16][17]. Since these compounds do not contain the naphthoquinone structure, these prodrugs may overcome the photoinstability and phototoxicity problems. It has been reported that the partition coefficient (log P) of MKH-DMG at physiological pH 7.4 is 3.66 [14]; this value indicates that MKH-DMG has good lipophilicity for membrane permeation. In addition, MKH-DMG and MKH-SUC could deliver MKH into hepatocellular carcinoma (HCC) cells more effectively than MK-4 [17]. Thus, MKH-DMG and MKH-SUC were considered to be good candidate compounds for skin application. In the current study, we examined their effectiveness as photostable and low-phototoxicity prodrugs of MKH in order to further development of a vitamin K skin application. The concept underlying the dermal delivery system of MKH is shown in Figure 1.

Photostability of MKH-Ester Derivatives in Artificial Sunlight
To evaluate the photostability of the MK-4 and MKH derivatives in skin application, 1 µM solutions of MK-4, MKH-DMG, or MKH-SUC in ethanol were irradiated with artificial sunlight (12,000 lx) at 25 • C with and without shading, and the final amounts of MKH derivatives and their monoesters, MK-4 and MK-4 chromenol, were determined by LC-MS/MS. The relative irradiance of artificial sunlight at wavelengths of 300 to 800 nm is almost equal to that of natural sunlight, according to the manufacturer. MK-4 and MKH-DMG concentrations decreased in sunlight irradiation according to an apparent first order rate (Figure 2A), but were unchanged under shading. MKH-SUC concentration decreased according to an apparent first order rate both with and without shading, although the rate of decrease was accelerated without shading. Apparent first order constants and half-lives are listed in Table 1. MKH-DMG and MKH-SUC were more photostable than MK-4 by ca. 50-and 3-fold, respectively. These results indicate that the use of MKH ester derivatives is a valid method to overcome the photolability of MK-4. In particular, MKH-DMG exhibits excellent photostability in potential skin application.
Following sunlight irradiation, MK-4 in ethanol was mainly converted to MK-4 chromenol. This photochemical result is consistent with previously reported results in polar solvents [11,12]. In contrast, MKH-DMG was mainly hydrolyzed to the corresponding monoester (MKH-mono-DMG). MKH-SUC was hydrolyzed to MKH-mono-succinate (MKH-mono-SUC) under shading, with this hydrolytic conversion being accelerated by unshaded sunlight irradiation ( Figure 2B) (Chromatograms of MK-4 chromenol, MKH-mono-DMG, and MKH-mono-SUC are shown in Supplemental Figure S1). These results suggest that the difference in the degree of hydrolysis of MKH-DMG and MKH-SUC at shading is probably due to the instability of the ester bonds.

Photostability of MKH-Ester Derivatives in Artificial Sunlight
To evaluate the photostability of the MK-4 and MKH derivatives in skin application, 1 μM solutions of MK-4, MKH-DMG, or MKH-SUC in ethanol were irradiated with artificial sunlight (12,000 lx) at 25 °C with and without shading, and the final amounts of MKH derivatives and their monoesters, MK-4 and MK-4 chromenol, were determined by LC-MS/MS. The relative irradiance of artificial sunlight at wavelengths of 300 to 800 nm is almost equal to that of natural sunlight, according to the manufacturer. MK-4 and MKH-DMG concentrations decreased in sunlight irradiation according to an apparent first order rate (Figure 2A), but were unchanged under shading. MKH-SUC concentration decreased according to an apparent first order rate both with and without shading, although the rate of decrease was accelerated without shading. Apparent first order constants and half-lives are listed in Table 1. MKH-DMG and MKH-SUC were more photostable than MK-4 by ca. 50-and 3-fold, respectively. These results indicate that the use of MKH ester derivatives is a valid method to overcome the photolability of MK-4. In particular, MKH-DMG exhibits excellent photostability in potential skin application.
Following sunlight irradiation, MK-4 in ethanol was mainly converted to MK-4 chromenol. This photochemical result is consistent with previously reported results in polar solvents [11,12]. In contrast, MKH-DMG was mainly hydrolyzed to the corresponding monoester (MKH-mono-DMG). MKH-SUC was hydrolyzed to MKH-mono-succinate (MKH-mono-SUC) under shading, with this hydrolytic conversion being accelerated by unshaded sunlight irradiation ( Figure 2B) (Chromatograms of MK-4 chromenol, MKH-mono-DMG, and MKH-mono-SUC are shown in Supplemental Figure S1). These results suggest that the difference in the degree of hydrolysis of MKH-DMG and MKH-SUC at shading is probably due to the instability of the ester bonds.    To evaluate the effect of irradiation wavelength on photostability of MK-4 and MKH-ester derivatives, 1 µM solutions of MK-4, MKH-DMG, and MKH-SUC in ethanol were irradiated with a xenon light source monochromatic light at 279, 341, 373, 404, or 435 nm, and the concentration of each compound was measured by LC-MS/MS. The log of concentration of each compound was plotted against irradiation energy (Figure 3), and the rates and half-life energies of degradation were obtained from slopes (Table 2). MK-4 decomposition, primarily due to MK-4 chromenol, was accelerated by a wide range of wavelengths (279−435 nm, Figure 3A). Hydrolytic degradations of MKH-DMG and MKH-SUC to their corresponding monoesters were accelerated only at the short wavelength of 279 nm ( Figure 3B,C). REVIEW 4 To evaluate the effect of irradiation wavelength on photostability of MK-4 and MKH-ester derivatives, 1 μM solutions of MK-4, MKH-DMG, and MKH-SUC in ethanol were irradiated with a xenon light source monochromatic light at 279, 341, 373, 404, or 435 nm, and the concentration of each compound was measured by LC-MS/MS. The log of concentration of each compound was plotted against irradiation energy (Figure 3), and the rates and half-life energies of degradation were obtained from slopes (Table 2). MK-4 decomposition, primarily due to MK-4 chromenol, was accelerated by a wide range of wavelengths (279−435 nm, Figure 3A). Hydrolytic degradations of MKH-DMG and MKH-SUC to their corresponding monoesters were accelerated only at the short wavelength of 279 nm ( Figure 3B,C).

Phototoxicity of MK-4 and MKH-Ester Derivatives
Some pharmaceutical substances are photosensitizers that can lead to the generation of reactive oxygen species (ROS) that cause oxidative damage to skin cells. It has been also shown that vitamin K 1 exhibits phototoxicity against cultured epidermis upon UVA irradiation [13]. Therefore, we evaluated the generation of ROS from MK-4 and MKH derivatives by UVA (320-400 nm) irradiation, which is strongly related to phototoxicity and accounts for approximately 95% of total ultraviolet radiation in sunlight that reaches the surface of the earth.
It has been reported that singlet oxygen promotes the peroxidation of skin surface lipids, resulting in the induction of skin inflammation [18]. Thus, we measured the irradiation intensity-dependent generation of singlet oxygen by MK-4, MKH-DMG, and MKH-SUC in phosphate-buffered aqueous solution after UVA irradiation using a p-nitrosodimethylaniline-based ROS assay ( Figure 4). Following UVA irradiation (0-15 J/cm 2 ), MK-4 exhibited increased formation of singlet oxygen in an irradiation energy-dependent manner similar to that of ketoprofen, the positive control, whereas MKH-DMG and MKH-SUC showed little generation of singlet oxygen, comparable to that of sulisobenzone, the negative control ( Figure 4). The results indicate that the quinone structure of MK-4 accelerates singlet oxygen generation by UVA irradiation and that the singlet oxygen formation of MKH ester derivatives is unaffected. The in vitro phototoxicity of MK-4 and MKH derivatives were assessed by measuring the intracellular ROS generation and the viability of human epidermal keratinocyte cells (HaCaT) after UVA irradiation in the presence or absence of MK-4 and MKH derivatives. Intracellular ROS generation was determined using the cell permeable probe 2′,7′-dichlorofluorescein diacetate (DCFH-DA), which rapidly reacts with hydroxyl radicals. Cells without drug treatment or UVA irradiation were used as the control. ROS generation and cell viability are shown in Figures 5 and 6, respectively.
None of the compounds increased intracellular ROS levels in the absence of UVA ( Figure 5). Following UVA irradiation, the intracellular ROS level in the MK-4 treated group was elevated in a dose-dependent manner, whereas no significant changes in intracellular ROS levels were observed in the MKH-DMG and MKH-SUC treated groups ( Figure 5). The in vitro phototoxicity of MK-4 and MKH derivatives were assessed by measuring the intracellular ROS generation and the viability of human epidermal keratinocyte cells (HaCaT) after UVA irradiation in the presence or absence of MK-4 and MKH derivatives. Intracellular ROS generation was determined using the cell permeable probe 2 ,7 -dichlorofluorescein diacetate (DCFH-DA), which rapidly reacts with hydroxyl radicals. Cells without drug treatment or UVA irradiation were used as the control. ROS generation and cell viability are shown in Figures 5 and 6, respectively.
None of the compounds increased intracellular ROS levels in the absence of UVA ( Figure 5). Following UVA irradiation, the intracellular ROS level in the MK-4 treated group was elevated in a dose-dependent manner, whereas no significant changes in intracellular ROS levels were observed in the MKH-DMG and MKH-SUC treated groups ( Figure 5).
None of the compounds affected cell viability in the absence of UVA irradiation ( Figure 6). After UVA irradiation, cells were re-incubated for 24 h and analyzed for cell viability. The viability of MK-4-treated cells was decreased dose-dependently (Figure 6), while the viability of cells treated with MKH-DMG or MKH-SUC was unaffected ( Figure 6). Increased intracellular ROS and decreased cell viability were only observed in cells treated with MK-4 and irradiated with UVA, suggesting that the phototoxicity of MK-4 was a result of ROS formation. Because DCFH-DA mainly reacts with highly reactive ROS such as hydroxyl radicals and peroxynitrate, cell toxicity may be caused by ROS and/or induced H 2 O 2 .
were used as the control. ROS generation and cell viability are shown in Figures 5 and 6, respectively.
None of the compounds increased intracellular ROS levels in the absence of UVA ( Figure 5). Following UVA irradiation, the intracellular ROS level in the MK-4 treated group was elevated in a dose-dependent manner, whereas no significant changes in intracellular ROS levels were observed in the MKH-DMG and MKH-SUC treated groups ( Figure 5).   Based on these results, it appears that MKH derivatives would be useful for skin application without producing singlet oxygen related skin surface toxicity and ROS related cytotoxicity induced by photo irradiation.

MKH Delivery into HaCaT Cells Using MK-4 and MKH-Ester Derivatives
The potential of MKH derivatives and MK-4 as dermal delivery systems of MKH were assessed by measuring intracellular MKH levels. However, accurate measurement of MKH is difficult because MKH is readily oxidized to MK-4. In γ-glutamyl carboxylation of vitamin K dependent protein (VKDP), MKH acts as a cofactor of γ-glutamyl carboxylase (GGCX) and is stoichiometrically converted to menaquinone-4 epoxide (MKO). Thus, MKO levels in HaCaT cells can reflect the amount of MKH delivered into the cell and subsequently converted in the carboxylation of VKDP. In addition, MK-4 level in cells after MKH prodrugs treatment indicates the oxidative product of MKH.
In this study, the intracellular concentration of MKO and MK-4 in HaCaT cells over time after treatment with MK-4, MKH-DMG, or MKH-SUC (5 μM) was determined by LC-MS/MS (Figure 7).
After MK-4 treatment, MKO and MK-4 levels immediately increased and reached a plateau after 12 to 24 h. Following MKH-DMG and MKH-SUC treatment, MKO levels were significantly increased, but the increase in MKO levels with MKH-SUC was faster than that with MKH-DMG. The areas under the concentration vs time curves over 72 h for MKO (AUCMKO (0-72 h)) and MK-4 (AUCMK-4 (0-72 h)) were calculated using the trapezoidal method and are shown in Table 3. The AUCMKO (0-72 h) for MKH-DMG and MKH-SUC was 1.023-and 1.641-fold that of MK-4, respectively. These results indicate that MKH-DMG and MKH-SUC can act as MKH prodrugs and effectively deliver MKH to keratinocytes.
The time course of intracellular MKO depends on the cellular uptake rate of prodrugs and on Based on these results, it appears that MKH derivatives would be useful for skin application without producing singlet oxygen related skin surface toxicity and ROS related cytotoxicity induced by photo irradiation.

MKH Delivery into HaCaT Cells Using MK-4 and MKH-Ester Derivatives
The potential of MKH derivatives and MK-4 as dermal delivery systems of MKH were assessed by measuring intracellular MKH levels. However, accurate measurement of MKH is difficult because MKH is readily oxidized to MK-4. In γ-glutamyl carboxylation of vitamin K dependent protein (VKDP), MKH acts as a cofactor of γ-glutamyl carboxylase (GGCX) and is stoichiometrically converted to menaquinone-4 epoxide (MKO). Thus, MKO levels in HaCaT cells can reflect the amount of MKH delivered into the cell and subsequently converted in the carboxylation of VKDP. In addition, MK-4 level in cells after MKH prodrugs treatment indicates the oxidative product of MKH.
In this study, the intracellular concentration of MKO and MK-4 in HaCaT cells over time after treatment with MK-4, MKH-DMG, or MKH-SUC (5 µM) was determined by LC-MS/MS (Figure 7).
The time course of intracellular MKO depends on the cellular uptake rate of prodrugs and on the conversion rate to MKH in cells. Intracellular delivery of MKH after MK-4 administration was dependent on the reductive activation of MK-4 to MKH. The ratio of AUC MKO /AUC MK-4 shown in Table 3 suggests the difference of delivery processes between MKH prodrug and MK-4. These ratios after MKH prodrugs administration were clearly higher than those obtained after MK-4 administration, indicating that these esters are hydrolyzed to MKH and effectively converted to MKO in HaCaT cells.  Table 3 suggests the difference of delivery processes between MKH prodrug and MK-4. These ratios after MKH prodrugs administration were clearly higher than those obtained after MK-4 administration, indicating that these esters are hydrolyzed to MKH and effectively converted to MKO in HaCaT cells.  To confirm whether MKH prodrugs could successfully deliver MKH to dermal cells under sunlight, the intracellular level of MKO in HaCaT cells was determined following administration of MK-4 and MKH derivatives in the presence or absence of artificial sunlight irradiation. The intracellular MKO level after administration of MK-4 and irradiation with artificial sunlight was seriously reduced compared to that following MK-4 administration in shade (Figure 8). MKO levels were increased after administration of MKH-DMG and MKH-SUC in sunlight, but these were approximately 60% that of the cells treated with their respective compounds in shading conditions. After MK-4 treatment, MKO and MK-4 levels immediately increased and reached a plateau after 12 to 24 h. Following MKH-DMG and MKH-SUC treatment, MKO levels were significantly increased, but the increase in MKO levels with MKH-SUC was faster than that with MKH-DMG. The areas under the concentration vs time curves over 72 h for MKO (AUC MKO (0-72 h) ) and MK-4 (AUC MK-4 (0-72 h) ) were calculated using the trapezoidal method and are shown in Table 3. The AUC MKO (0-72 h) for MKH-DMG and MKH-SUC was 1.023-and 1.641-fold that of MK-4, respectively. These results indicate that MKH-DMG and MKH-SUC can act as MKH prodrugs and effectively deliver MKH to keratinocytes.   To confirm whether MKH prodrugs could successfully deliver MKH to dermal cells under sunlight, the intracellular level of MKO in HaCaT cells was determined following administration of MK-4 and MKH derivatives in the presence or absence of artificial sunlight irradiation. The intracellular MKO level after administration of MK-4 and irradiation with artificial sunlight was seriously reduced compared to that following MK-4 administration in shade (Figure 8). MKO levels were increased after administration of MKH-DMG and MKH-SUC in sunlight, but these were approximately 60% that of the cells treated with their respective compounds in shading conditions. These results clearly indicate that the MKH prodrugs (MKH-DMG and MKH-SUC) could deliver MKH into keratinocytes even under sunlight irradiation, but MK-4 could not. It is also possible that MK-4 chromenol, the photodegradation product of MK-4, could not reconvert to MK-4 or MKH in keratinocytes.
Int. J. Mol. Sci. 2019, 20 FOR PEER REVIEW 9 These results clearly indicate that the MKH prodrugs (MKH-DMG and MKH-SUC) could deliver MKH into keratinocytes even under sunlight irradiation, but MK-4 could not. It is also possible that MK-4 chromenol, the photodegradation product of MK-4, could not reconvert to MK-4 or MKH in keratinocytes.

Discussion
The relative irradiance of artificial sunlight used in this study at wavelengths of 300 to 800 nm is approximately equal to that of natural sunlight. Thus, the high photostability of MKH derivatives compared with the quinone form of MK-4 suggests that MKH derivatives may be useful for skin application without photodegradation. A study of the vitamin K1 photoreaction by Hangarter et al. showed that charge transfer from the β,γ-double bond of the isoprenyl side chain to the quinone moiety initiates intramolecular proton transfer from the side chain, yielding the 1,3-quinone methide diradical in polar solvents and subsequently forms 1,2-quinone methide, which forms vitamin K1 chromenol [12]. The formation of MK-4 chromenol in ethanol following irradiation of MK-4 suggests that it is formed by the same mechanism. The bis-ester of MKH was hydrolyzed to MKH monoester by irradiation, which may also be explained by the formation of 1,3-quinone methide in a polar solvent that could then intramolecularly transfer from the side chain to the quinone moiety. MKH-DMG and MKH-SUC hydrolyses were accelerated only at 279 nm, confirming the hydrolyses of MKH-bis-ester to monoester in sunlight.
MK-4 degraded following a reaction rate-light wavelength relationship, in which exposure to shorter wavelengths resulted in faster degradation rates. However, Teraoka et al. showed that the curve of photodegradation rate of MK-4 against wavelength followed a bell bottom shape and that the fastest rate occurred at 430 nm [10]. In their study, MK-4 was solubilized in rapeseed oil, which shows large absorptivity at <300 nm. Thus, it is likely that irradiation energy at <300 nm was absorbed by rapeseed oil, resulting in reduced exposure of MK-4 to irradiation energy and producing the bell bottom profile.
Our results indicate that UVA irradiation of MK-4 accelerates singlet oxygen generation in aqueous solution, and increases intracellular ROS generation and cell toxicity in the HaCaT keratinocyte cell line. These results may be related to 1,3-quinone methide diradical formation by UVA irradiation of MK-4. Contrastingly, MKH derivatives avoided ROS formation and phototoxicity under UVA irradiation. Therefore, MKH derivatives would be candidates for effective, nonphototoxic use in skin application.
In order to act as an MKH prodrug, MKH derivatives must be converted to the parent drug in

Discussion
The relative irradiance of artificial sunlight used in this study at wavelengths of 300 to 800 nm is approximately equal to that of natural sunlight. Thus, the high photostability of MKH derivatives compared with the quinone form of MK-4 suggests that MKH derivatives may be useful for skin application without photodegradation. A study of the vitamin K 1 photoreaction by Hangarter et al. showed that charge transfer from the β,γ-double bond of the isoprenyl side chain to the quinone moiety initiates intramolecular proton transfer from the side chain, yielding the 1,3-quinone methide diradical in polar solvents and subsequently forms 1,2-quinone methide, which forms vitamin K 1 chromenol [12]. The formation of MK-4 chromenol in ethanol following irradiation of MK-4 suggests that it is formed by the same mechanism. The bis-ester of MKH was hydrolyzed to MKH monoester by irradiation, which may also be explained by the formation of 1,3-quinone methide in a polar solvent that could then intramolecularly transfer from the side chain to the quinone moiety. MKH-DMG and MKH-SUC hydrolyses were accelerated only at 279 nm, confirming the hydrolyses of MKH-bis-ester to monoester in sunlight.
MK-4 degraded following a reaction rate-light wavelength relationship, in which exposure to shorter wavelengths resulted in faster degradation rates. However, Teraoka et al. showed that the curve of photodegradation rate of MK-4 against wavelength followed a bell bottom shape and that the fastest rate occurred at 430 nm [10]. In their study, MK-4 was solubilized in rapeseed oil, which shows large absorptivity at <300 nm. Thus, it is likely that irradiation energy at <300 nm was absorbed by rapeseed oil, resulting in reduced exposure of MK-4 to irradiation energy and producing the bell bottom profile.
Our results indicate that UVA irradiation of MK-4 accelerates singlet oxygen generation in aqueous solution, and increases intracellular ROS generation and cell toxicity in the HaCaT keratinocyte cell line. These results may be related to 1,3-quinone methide diradical formation by UVA irradiation of MK-4. Contrastingly, MKH derivatives avoided ROS formation and phototoxicity under UVA irradiation. Therefore, MKH derivatives would be candidates for effective, non-phototoxic use in skin application.
In order to act as an MKH prodrug, MKH derivatives must be converted to the parent drug in dermal cells after skin application. Our results clearly show that MKH derivatives and MK-4 increase intracellular MKO levels in the keratinocyte cell line we used, indicating that they could be used to deliver MKH into keratinocytes, where it then acts as a cofactor for GGCX. We previously showed that MKH-DMG and MKH-SUC reconvert to MKH in a reaction catalyzed by carboxyl esterase [17], which is present in dermal cells [19][20][21], so accelerated intracellular reconversion to a parent drug is possible. Although UVA-irradiated MK-4 did not increase intracellular MKO, UVA-irradiated MKH derivatives were able to.
MKH delivery with MKH-SUC into HaCaT cells is faster than that with MKH-DMG. A similar result was obtained while using hepatocellular carcinoma (HCC) cells in a previous study-the uptake and reconversion rates of MKH-SUC in HCC cells were faster than those of MKH-DMG [17]. Therefore, it can be presumed that the rapid and increased MKH delivery with MKH-SUC in HaCaT cells is due to the rapid uptake and fast reconversion rates.
In conclusion, MKH-DMG and MKH-SUC could act as MKH prodrugs for skin application without photoinstability and phototoxicity. This ideal strategy is a safer and more efficient way to deliver MKH than using the quinone form MK-4. Therefore, MKH prodrugs have the potential to be used for skin diseases which require the activity of vitamin K dependent proteins for treatment, even under direct sunlight irradiation. Further studies that can reflect the actual skin environment using human origin skin models such as 3-dimensional skin models or flesh skin tissue samples are needed.

Cell Culture
Human epidermal keratinocyte cell line HaCaT was obtained from CLS Cell Lines Service GmbH (Eppelheim, Germany). Cells were maintained in DMEM (high glucose, CLS) with 10% FBS (Life Technologies, Carlsbad, CA, USA) and 1% penicillin/streptomycin (Life Technologies) at 37 • C under an atmosphere of 5% CO 2 .