The characteristics of aromatase deficient hairless mice indicate important roles of extragonadal estrogen in the skin

Available online at www.sciencedirect.com

Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

The characteristics of aromatase deficient hairless mice indicate important roles of extragonadal estrogen in the skin Kazue Tsukahara a , Shingo Kakuo a , Shigeru Moriwaki a , Mitsuyuki Hotta a , Atsushi Ohuchi a , Takashi Kitahara a , Nobuhiro Harada b,∗ b

a Biological Science Laboratories, Kao Corporation, Ichikai, Haga, Tochigi 321-3497, Japan Department of Biochemistry, Fujita Health University School of Medicine, Kutsukake-cho, Toyoake, Aichi 470-1192, Japan

Received 9 January 2007; accepted 12 July 2007

Abstract The roles of extragonadal estrogen in the skin are poorly understood, due to the lack of proper animal models. We examined the skin phenotypes of aromatase-knockout hairless (ArKO) mice and wild-type hairless (WT) mice, both of which were obtained through crossbreeding of Ar± mice and hairless mice. Differences in the skins of ArKO and WT mice were compared with those of ovariectomized (OVX) and control (Sham) mice. A difference was observed in the skin tone of ArKO mice, which is pale white and differs from the pinkish tone of all other mice. However, both ArKO and OVX mice similarly exhibited deteriorations of skin properties as compared to their respective controls. Furthermore, all the deteriorations were similarly amplified by chronic UVB irradiation in both ArKO and OVX mice as compared to their respective controls. The unique skin phenotype of ArKO mice was observed in sunburn reactions. Specifically, skins of ArKO mice showed no reaction after an acute UVB irradiation at dose intensities caused sunburn in others. However, follow-up observation found delayed reactions associated with brownish skin color and swelling only in ArKO mice, thereby suggesting that the role of extragonadal estrogen may be connected with the protective reactions of skin. © 2007 Elsevier Ltd. All rights reserved. Keywords: Aromatase; Ovariectomy; Estrogen; Photoaging; Wrinkle; Ultraviolet radiation

1. Introduction Estrogens are known to play a key role in reproductive functions and in physiological functions such as bone metabolism, the cardiovascular system and the central nervous system. The effects of estrogens on skin, which is the focus of this study, have also been identified, conventionally through studies of the various aging phenomena and skin problems such as wrinkles and sags which are associated with postmenopausal women. These studies have in turn reported the effectiveness of estrogen when administered postmenopausally in reducing skin-aging effects [1,2].

Abbreviations: ArKO, aromatase-knockout; MED, minimal erythemal dose; TEWL, transepidermal water loss; UVB, ultraviolet-B. ∗ Corresponding author. Tel.: +81 562 93 2450; fax: +81 562 93 1193. E-mail address: [email protected] (N. Harada). 0960-0760/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2007.07.004

The common method for identifying the effects of estrogen on skin is the surgical suppression in animal models of the major source of estrogen. Previously, we have determined the effectiveness of ovariectomized mice for studies of photoaging in skin [3]. Recently, aromatase (cytochrome P-450 AROM)-knockout mice have been proving effective as a way of analyzing the action of estrogen, as reported by separate groups through observations of disturbances both in sexual patterns and in the normal growth of genital organs [4–6]. However, researchers in the fields of dermatology have not yet evaluated the effectiveness of ArKO models for studying the extragonadal estrogen biosynthesized in many sites throughout the body (adipose tissue, bone, brain), despite being aware of extragonadal estrogen’s importance in the physiology and pathophysiology of postmenopausal women [7]. This study began with the confirmation of the sufficiency of ovariectomized HR/ICR (OVX) mice in comparison with

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

those of sham-operated controls (Sham). The lack of major estrogen was confirmed to accelerate aging of the skin. When we started crossing aromatase semi-deficient (Ar±) mice and Skh-1 albino hairless mice to create aromatase-knockout hairless (ArKO) mice, we had predicted deteriorations similar to those observed in the OVX mice. On the other hand, given that ovariectomy alone greatly accelerated skin aging, we also predicted that the amount of locally produced estrogen would be low, and that its impact on skin, if any, would be correspondingly small. However, from this study, we found that all ArKO mice possessed a distinctly pale white skin tone, whereas other offspring of this strain displayed pinkish skin tones similar to those of Sham and OVX mice. It is known that skin tones relate closely to UV sensitivity and other physiological functions of skin. Therefore, this unexpected difference in the skin tone of ArKO mice warrants further evaluation through comparison with WT mice with and without ovaries. Finally, we ought to point out that the ArKO and OVX models in this study were different strains. This difference was due in part to the animal ethics committee, which declined a request for the additional breeding of ArKO mice in this study on grounds of the low yield in the hetero breeding. It may not be meaningful to compare OVX with ArKO or to compare Sham with WT. As an alternative, the results from the OVX vs. Sham comparison were compared with those in ArKO and WT mice. More specifically, morphological and functional changes that have been observed between OVX and Sham (HR/ICR) were compared with those observed between ArKO and WT hairless (Skh-1) mice so as to determine the roles of extragonadal estrogen in skins.

2. Materials and methods According to the Japanese Law for the Humane Treatment and Management of Animals, the Kao Corporation Animal Ethics Committee approved all experiments described in this paper. 2.1. Animals 2.1.1. Aromatase-knockout (ArKO) and wild-type (WT) hairless mice Female Ar± mice [6] were matched with male Skh-1-hr BR-hairless mice (purchased from Charles River Laboratories, Kanagawa, Japan). The genotype of each mouse was determined from tail biopsies as described by Honda et al. [6]. Through four generations of selective breeding, 25 female ArKO and 25 female WT hairless mice were obtained and were used in this study. 2.1.2. Ovariectomized (OVX) and sham-operated (Sham) HR/ICR mice Female HR/ICR hairless mice were derived by crossing hairless mice (HR/HR), originally obtained from Nis-

83

seiken Corp. (Tokyo, Japan), with albino HaM/ICR mice. The HR/ICR strain represents a line maintained under clean conventional conditions in our laboratory by hairless brother/haired sister mating for several years. Fifty mice underwent either ovariectomy or sham operations. For ovariectomy, the bilateral flanks were opened, and the ovaries were carefully excised after ligation of the oviducts with catgut (Matsuda Sutures Catgut Plain 3-0, Matsuda Ika Kogyo Co., Ltd., Tokyo, Japan). The fascia was then sutured once, and the skin was sutured twice. For the sham operation, a similar procedure was performed except for the ligation of the oviducts and the excision of the ovaries. In all mice, 30 ␮l (100 mg/ml) KANAMYCIN Sulfate (100 mg/ml, Banyu Pharmaceutical Co., Ltd., Tokyo, Japan) was injected intramuscularly for 4 days after the operation, and 3 days after that, UVB irradiations were initiated. 2.2. The short-time irradiation of UVB 2.2.1. Minimal erythema dose (MED) The MED was measured on the dorsal skin of all mice (n = 9 in each group). Under pentobarbital anesthesia, the back of each mouse was flattened, covered with a masking template with four opening squares (10 mm × 6 mm), and irradiated with different dose intensities (from 10 to 90 mJ/cm2 ). The irradiation time was adjusted from 10 to 200 s in order to keep the irradiation densities. Among these 36 test-sites in each type of mouse, 4 test-sites were irradiated with either 10, 60, 70 or 80 mJ/cm2 and 5 test-sites were irradiated with either 20, 30, 40 or 50 mJ/cm2 using five Toshiba SE lamps (UVB) from a height of 30 cm [3,8]. These intensities were confirmed by a UV-radiometer 305/365D (Topcon Co., Ltd., Tokyo, Japan). 2.2.2. Erythema score (follow-up observation) Various signs of sunburn, e.g., erythema (redness), edema (inflammatory swelling) and blistering are known to occur 1–6 h after irradiation, peak at 12–24 h, and disappear around 4–7 days. In this study, five test-sites irradiated at 40 mJ/cm2 on the dorsal skins of ArKO mice were observed for 6 days in comparison with the same sites on WT mice. The time-course of reactions were graded according to the scale of Parrish et al. [9] as follows: 0 1 2 3 4 5

no response erythema (redness of sunburn) more than trace, filling most but not all of the exposure site, but with indefinite borders minimal erythema with four definite borders more pronounced erythema erythema with edema erythema with edema and vesiculation (blister)

2.3. The long-time irradiation of UVB According to the previous study, which determined that 6 weeks of UVB irradiation at dose intensities causing mini-

84

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

mum erythema induces wrinkle formation on the dorsal skin of mice [10], 64 mice (n = 16 in each group) were divided into two groups—half were irradiated with UVB for 6 weeks, and the other half were not. Intensities of UVB were set at 50 mJ/cm2 /day for the first week, increased 5 mJ/cm2 per week until week 4, and then kept at 65 mJ/cm2 /day for the remaining period. The UVB source was a bank of six Toshiba SE lamps without any filtering for UVB (peak of emission near 312 nm; irradiance between 290 and 320 nm corresponds to 55% of the total amount of UVB). The distance from the lamps to the animals’ backs was 35 cm. The energy output of the lamps was measured with a UV-radiometer 305/365D. The spectral irradiance of the lamps was measured with a MSR7000 radiospectrometer (Optical Science Co., Ltd., Tokyo, Japan). During this period, mice could move around freely in their cages. The duration and the intensities of irradiation were adequately adjusted by considering their susceptibility to sunburn. Photographs of the dorsal skin of each mouse were taken using a Minolta ␣707si camera with a macro 100 lens (Minolta, Co., Ltd., Tokyo, Japan) system. 2.3.1. Wrinkle quantification The increase of wrinkles caused by the long-term UVB irradiation was quantitatively evaluated using replica image analysis. Replicas were obtained from the back of each mouse using a rubber precision impression material (hydrophilic vinyl silicon impression material: GC Exafine, GC Co., Ltd., Tokyo, Japan) and were cut into circular pieces and placed on flat plates. The oblique lighting of a negative replica cast shadows of wrinkles and skin textures. The images of shadows were captured using a CCD, and the effects of skin textures were excluded using filtration techniques such as edge processing and binarization. The area ratio of the remaining shadows was calculated as the ratio of wrinkles. A PIAS LA-555 personal image analysis system (PIAS Corporation, Osaka, Japan) was used for this analysis, and our previous report provides more details [3].

2.3.4. Transepidermal water loss (TEWL) Cutaneous water evaporation was measured by using a Tewameter TM210® (Courage + Khazaka electronic GmbH, K¨oln, Germany) as previously described [11]. Measurements were performed once each on the right and left sides of the midline, and mean values were obtained. 2.3.5. Skin thickness The B-mode ultrasonographic image was obtained from each mouse five times in a dynamic range of 60 with a gain of 8 dB at 30 MHz using an UX-02 ultrasonic diagnostic system (RION Co., Ltd., Tokyo, Japan) as previously described [3,11]. The discontinuity of echogenicity on the image is defined as the border between the dermis and the subcutaneous tissue [12]. The thickness from the skin surface to the border was measured at 10 sites on each print, and mean values were calculated. 2.3.6. Skin elasticity Skin elasticity was measured by using a Cutometer Skin Elasticity Meter 575® (Courage + Khazaka electronic GmbH, K¨oln, Germany) as previously described [3,11]. This instrument depends on the principle of suction elongation described by Elsner et al. [13]. Briefly, the time/strain mode was used with application of a 100 mb load for 1 s followed by 1 s of relaxation. The skin deformation was then plotted as a function of time. The parameter used was immediate distension (Ue), measured at 0.1 s, as described by Agache et al. [14]. 2.4. Statistics Results are expressed as means ± standard deviation or means ± standard error, as noted in the figures and tables. Differences between two groups were analyzed using Student’s t-test, and differences among more than two groups

2.3.2. Skin color measurement The skin color of the midline (1 cm above the root of the tail) was measured five times using a color meter (Model OFC-300A, Nippon Denshoku Industries Co., Ltd., Tokyo, Japan) according to a previously described method [11]. The L* value (luminance, which represents the relative brightness from total darkness (L* = 0) to absolute white (L* = 100)), the a* value (red–green color axis) and the b* value (yellow–blue color axis) were calculated, and mean values were obtained. 2.3.3. Skin conductance in the stratum corneum measurement The conductance in the stratum corneum of the skin was assessed to detect the functional deterioration caused by chronic UVB irradiation, a Skicon-200® (IBS Co., Ltd., Hamamatsu, Japan) used as previously described [11]. Measurements were performed five times each on the right and left of the midline, and mean values were obtained.

Fig. 1. The constitutional skin tones of female ArKO mice (left) and of WT mice (right) at 8 week of age in natural condition.

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

were analyzed using two-way ANOVA, followed by Tukey’s post hoc multiple comparison test. Statistical significance was declared for differences with p < 0.05.

85

tone and in body size. This photo was taken when mice were in natural conditions. Fig. 2A and B show the skin surface morphology of hairless mice with or without chronic UVB irradiation for 6 weeks. These photos were taken when mice were under anesthesia.

3. Results 3.2. Effects of the short-time UVB 3.1. Weight and external appearance of animals Significant gain in the mean weight of ArKO and OVX mice (p < 0.01 and 0.05, respectively) were observed as compared to their respective controls: ArKO 32.3 g, WT 28.6 g, OVX 30.2 g, Sham 28.6 g. Fig. 1 shows the difference between ArKO and WT mouse in constitutional skin

3.2.1. Minimal erythema dose The MED (minimal UVB dose to generate erythema) was 39.3 mJ/cm2 in the Sham mice and 27.2 mJ/cm2 in the OVX mice, showing that ovariectomy results in erythema formation at a significantly lower UVB dose. In WT mice, the MED was 42.1 mJ/cm2 , similar to that of the Sham mice. In con-

Fig. 2. Skin features of mice after chronic UVB irradiation (five times weekly for 6 weeks). (A) WT mice and ArKO mice. (B) Sham-operated HR/ICR mice and OVX HR/ICR mice.

86

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

Fig. 3. Minimal erythema dose (MED) values of mice. The amount of UVB radiation, measured in energy per unit area (mJ/cm2 ), to produce erythema at an exposed site is defined as the MED. N = 9. N.D., not detectable.

trast, ArKO mice showed no erythema, even at a UVB dose of 90 mJ/cm2 (Fig. 3), indicating an unexpected tolerance to sunburn. 3.2.2. Erythema score The unexpected tolerance of ArKO mice to sunburn was examined in a subsequent experiment on test-sites irradiated at 40 mJ/cm2 in comparison with WT mice. The skins of WT mice appeared to recover completely from sunburn by day 6, In contrast, the skins of ArKO mice became brown and their surfaces swelled, showing signs of inflammatory edema on day 3. The brownness and edema peaked on day 4, and faded thereafter. However, according to the scale of Parrish et al. [9], this inflammatory reaction with no associated erythema (redness) was scored as 0 = no response (Fig. 4). 3.3. Effect of chronic UVB 3.3.1. Wrinkle formation Fig. 5A shows wrinkles on the dorsal skins of mice after 6 weeks of UVB irradiation. UVB irradiation increased wrinkles both in ArKO and in OVX mice more than in their respective controls. Those increases might not be easily detectable in this figure. Fig. 5B shows quantified ratios of those wrinkles, confirming that both the ArKO and the OVX mice had significantly more wrinkles than their respective controls, and that the ArKO mice had significantly more wrinkles than the OVX mice.

Fig. 4. Erythema scores of ArKO and WT mice. Using the six-point scale of Parish et al. [9], the erythema scores of ArKO mice were measured in the five test-sites irradiated at 40 mJ/cm2 in comparison with WT mice for 6 days.

3.3.2. Skin color Table 1A shows skin color measurements of Ar mice with or without UVB irradiation. (1) Without UVB, ArKO mice had a significantly higher L* (brightness) value, and significantly lower a* (redness) and b* (yellowness) values compared to WT mice. (2) With UVB, ArKO mice had a significantly higher L* value, a significantly lower a* value, but no significant difference in b* value compared to WT mice. (3) UVB irradiated WT mice had significantly higher values for a* , b* and L* compared to non-irradiated WT mice. (4) UVB irradiated ArKO mice had significantly higher a* and b* values, but no significant difference in L value compared to non-irradiated ArKO mice. Table 1B shows skin color measurements for HR/ICR mice with or without UVB irradiation. (1) Without UVB, OVX mice had no difference in L* , a* or b* values compared to the Sham mice. (2) With UVB, OVX mice had a significantly higher L* value, but no significant difference in either a* or b* values compared to the Sham mice. (3) UVB irradiated Sham mice had significantly higher a* and b* values, but no significant difference in L* value compared to non-irradiated Sham mice. (4) UVB irradiated OVX mice had a significantly higher a* and b* value, but no significant difference in L* values compared to non-irradiated OVX mice.

Table 1A Skin color in ArKO or in WT mouse skin, with or without chronic UVB irradiation for 6 weeks (mean ± S.D.) Parameter

L* a* b*

UVB(−)

UVB(+)

WT

ArKO

WT

ArKO

48.66 ± 1.84 0.792 ± 0.29 −1.670 ± 0.89

56.57 ± 1.90A −0.275 ± 0.60A −7.499 ± 0.59A

51.67 ± 0.82a 1.514 ± 0.38a 2.961 ± 1.13A

57.59 ± 1.69C 0.707 ± 0.21Bc 3.290 ± 0.65B

Letters in superscript indicate: (A, a) p < 0.01, 0.05 compared with UVB(−)/WT, respectively. (B) p < 0.01 compared with UVB(−)/ArKO. (C, c) p < 0.01, 0.05 compared with UVB(+)/WT, respectively. N = 8 per group.

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

87

Fig. 5. (A) Wrinkles of mice after chronic UVB irradiation. The skin is illuminated from an oblique light source. (B) The area ratio of UVB-induced wrinkles on the dorsal skin of mice. The ratio was calculated using replica image analysis. N = 8. * p < 0.05 compared with OVX mice.

Table 1B Skin color in OVX or in sham-operated HR/ICR mouse skin, with or without chronic UVB irradiation for 6 weeks (mean ± S.D.) Parameter

L* a* b*

UVB(−)

UVB(+)

Sham

OVX

Sham

OVX

45.15 ± 1.13 1.202 ± 0.31 −2.480 ± 0.64

46.77 ± 1.51 1.192 ± 0.12 −2.515 ± 0.29

45.05 ± 0.62 2.104 ± 0.50a 0.908 ± 0.90A

47.26 ± 0.59c 2.206 ± 0.51B 0.847 ± 1.18B

Letters in superscript indicate: (A, a) p < 0.01, 0.05 compared with UVB(−)/Sham, respectively. (B) p < 0.01 compared with UVB(−)/OVX. (c) p < 0.05 compared with/UVB(+)/Sham. N = 8 per group

3.3.3. Skin conductance of the stratum corneum (1) Ar strain: The skin conductance of the stratum corneum did not differ between ArKO and WT mice without UVB irradiation. Chronic UVB irradiation significantly decreased skin conductance in both types of mice. The decrease was more significant in the ArKO mice than in the WT mice (Table 2A). (2) HR/ICR strain: The skin conductance of the stratum corneum did not differ between OVX and Sham mice without UVB irradiation. Chronic UVB irradiation sig-

nificantly decreased skin conductance in both types of mice. The decrease in the OVX mice was more than in the Sham mice, but the difference was insignificant (Table 2B). 3.3.4. Transepidermal water loss (TEWL) (1) Ar strain: The TEWL did not differ between the ArKO and the WT mice without UVB irradiation. Chronic UVB irradiation increased TEWL in both types of mice. The increase was more significant in the ArKO mice than in the WT mice (Table 2A).

88

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

Table 2A Skin properties in ArKO or in WT mouse skin, with or without chronic UVB irradiation for 6 weeks (mean ± S.D.) UVB(−)

UVB(+)

WT (␮−1 )

Skin conductance TEWL (␮g/cm2 /h) Skin thickness (mm) Skin elasticity (Ue, mm)

474.4 3.49 0.40 0.24

ArKO ± ± ± ±

75.3 0.74 0.03 0.05

447.2 3.73 0.50 0.14

WT ± ± ± ±

106.0 0.67 0.01a 0.02A

ArKO

200.5 8.55 0.66 0.08

± ± 2.47A ± 0.07A ± 0.02A 15.7A

64.7 12.9 0.75 0.05

± ± ± ±

18.4Bc 3.17BC 0.05Bc 0.00B

Letters in superscript indicate: (A, a) p < 0.01, 0.05 compared with UVB(−)/WT, respectively. (B) p < 0.01 compared with UVB(−)/ArKO. (C, c) p < 0.01, 0.05 compared with UVB(+)/WT, respectively. N = 8 per group. Table 2B Skin properties in OVX or in sham-operated HR/ICR mouse skin, with or without chronic UVB irradiation for 6 weeks (mean ± S.D.) UVB(−)

UVB(+)

Sham (␮−1 )

Skin conductance TEWL (␮g/cm2 /h) Skin thickness (mm) Skin elasticity (Ue, mm)

463.0 5.56 0.36 0.31

OVX ± ± ± ±

123 0.94 0.02 0.02

438.8 5.61 0.45 0.18

Sham ± ± ± ±

62.9 0.16 0.01A 0.03A

152.5 12.59 0.50 0.08

OVX ± ± 3.38a ± 0.01A ± 0.02A 38.8A

75.4 20.07 0.63 0.04

± ± ± ±

22.6B 5.88Bc 0.04BC 0.01B

Letters in superscript indicate: (A, a) p < 0.01, 0.05 compared with UVB(−)/Sham, respectively. (B) p < 0.01 compared with UVB(−)/OVX. (C, c) p < 0.01, 0.05 compared with UVB(+)/Sham, respectively. N = 8 per group.

(2) HR/ICR strain: The TEWL did not differ between the OVX and the Sham mice without UVB irradiation. Chronic UVB irradiation increased TEWL in both types of mice. The increase was more significant in the OVX mice than in the Sham mice (Table 2B). 3.3.5. Skin thickness (1) Ar strain: The skin thickness was significantly greater in the ArKO mice than in the WT mice with or without chronic UVB irradiation. Chronic UVB significantly increased the skin thickness in both types of mice (Table 2A). (2) HR/ICR strain: The skin thickness was significantly greater in the OVX mice than in the Sham mice with or without chronic UVB irradiation. Chronic UVB significantly increased the skin thickness in both types of mice (Table 2B). 3.3.6. Skin elasticity (1) Ar strain: The skin elasticity (Ue) was significantly lower in the ArKO mice than in the WT mice without UVB irradiation. Chronic UVB irradiation significantly decreased the skin elasticity in both types of mice. The ArKO mice showed a tendency towards a decreased Ue compared with the WT mice, but it was not statistically significant (Table 2A). (2) HR/ICR strain: The skin elasticity was significantly lower in the OVX mice without UVB irradiation. Chronic UVB irradiation decreased skin elasticity in both types of mice. The OVX mice showed a tendency toward a decreased Ue compared with the Sham mice, but it was not statistically significant (Table 2B).

4. Discussion In terms of the barrier functions of the skin, ArKO mice showed deterioration compared with WT mice. The range of difference was similar to that between OVX and Sham mice. Miyauchi et al. [15] have reported a marked decrease in skin surface water content and water retention ability after UVB irradiation of guinea pig skin, while Haratake et al. [16] have reported an increase in TEWL of the skin of hairless mice after UVB irradiation. Those previous findings suggest that the functional deteriorations are amplified by UVB irradiation and thereby elicit potential differences other than in skin tones. Further deteriorations in terms of TEWL and skin conductance were also observed after 6 weeks of irradiation with UVB. However, the ArKO mice exhibited a similar level of deterioration to the OVX mice compared to their respective controls. In other words, stratum corneum functions might not be affected by extragonadal estrogen. Estrogen has been reported to increase hyaluronic acid synthesis [17] and to promote collagen synthesis [18]. In our own previous study, which examined the plantar skin of OVX and Sham rats chronically irradiated with low-dose UVB, the OVX rats showed more wrinkle formation, larger decreases in skin elasticity and an earlier stage of curling in the three-dimensional structure of their dermal elastic fibers as compared to the Sham rats [19]. The present study too observed more wrinkle formation both in the ArKO and in the OVX mice as compared to their respective controls. In addition, the ArKO mice had significantly more wrinkles compared to the OVX mice, thereby indicating that the low levels of estrogen that might be remaining in the OVX mice had suppressed their wrinkle formation. Given these findings, we speculate that estrogen plays an important role in

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

the homeostasis of the dermal extracellular matrix, and thus the lack of estrogen in OVX mice as well as the complete absence of estrogen in the ArKO mice leads to a loss of the homeostatic control, thereby accelerating skin aging. In terms of skin color measurements, the redness of ArKO mice in this study was significantly lower than the WT mice, suggesting that the ArKO mice had less blood flow in the skin. Unexpectedly, no difference of the redness was practically observed between OVX and the Sham mice. Previous reports of the effects of estrogen on vascular cells [20–22] and on cutaneous blood flow distribution [23,24] suggest that estrogen has a physiological role in the blood circulatory system. The clinical effects of hormone replacement therapy on blood flow velocity [25] and on hemoglobin and oxidized hemoglobin levels [26] also suggest that levels of estrogen are important in the circulatory system. Hiramoto et al. reported that the degree of UV-induced edema is higher in the ears of male mice compared to female mice, and that intraperitoneal injections of 17␤-estradiol prevented the development of edema in the ears of male mice [27]. Our own observation shows more erythema (redness) in the OVX mice compared to the Sham mice. Surprisingly, no erythema was observed in ArKO mice in a short-term observation after an acute dose of UVB, whereas erythema similar to that in the Sham mice was observed in the WT mice. This short-term observation indicates that the ArKO mice might have higher tolerance to sunburn; however, follow-up observations of the same mice found delayed inflammatory reactions associated with brownish skin color and swelling. Assuming that absence of local estrogen in ArKO mice may have an impact on cutaneous blood flow which in turn changes inflammatory reaction, we conducted a preliminary experiment that topically applied 17␤-estradiol on the dorsal skin of ArKO mice for 3 weeks prior to a single UVB irradiation. As a result, the abnormal inflammatory reaction of ArKO mice was well rescued by this treatment (data not shown), thereby suggesting the importance of estradiol in skin. In summary, the changes in the skin of ArKO mice and of OVX mice from their respective controls were similar in terms of conductance, thickness, elasticity and TEWL. In contrast, they differed from their respective controls with respect to skin tone, UV sensitivity and wrinkle ratio. Throughout this study, the skin of WT mice behaved similar to that of Sham mice and thus allowed us to conclude that the effect of strain differences is small and that extragonadal estrogen is the most probable cause of these differences. An experimental approach using ArKO mice may provide a new avenue for the systematic evaluation of the roles of estrogens in the skin.

[10]

References

[21]

[1] P. Creidi, B. Faivre, P. Agache, E. Richard, V. Haudiquet, J.P. Sauvanet, Effect of a conjugated oestrogen (Premarin® ) cream on ageing facial

[2]

[3]

[4]

[5]

[6]

[7] [8]

[9]

[11]

[12]

[13] [14]

[15]

[16]

[17] [18]

[19]

[20]

89

skin. A comparative study with a placebo cream, Maturitas 19 (1994) 211–223. L.B. Dunn, M. Damesyn, A.A. Moore, D.B. Reuben, G.A. Greendale, Does estrogen prevent skin aging? Results from the First National Health and Nutrition Examination Survey (NHANES I), Arch. Dermatol. 133 (1997) 339–342. K. Tsukahara, H. Nakagawa, S. Moriwaki, S. Kakuo, A. Ohuchi, Y. Takema, G. Imokawa, Ovariectomy is sufficient to accelerate spontaneous skin ageing and to stimulate ultraviolet irradiation-induced photoageing of murine skin, Br. J. Dermatol. 151 (2004) 984– 994. C.R. Fisher, K.H. Graves, A.F. Parlow, E.R. Simpson, Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 6965– 6970. K.M. Robertson, L. O’Donnell, M.E. Jones, S.J. Meachem, W.C. Boon, C.R. Fisher, K.H. Graves, R.I. McLachlan, E.R. Simpson, Impairment of spermatogenesis in mice lacking a functional aromatase (cyp 19) gene, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 7986–7991. S. Honda, N. Harada, S. Ito, Y. Takagi, S. Maeda, Disruption of sexual behavior in male aromatase-deficient mice lacking exons 1 and 2 of the cyp 19 gene, Biochem. Biophys. Res. Commun. 252 (1998) 445–449. E.R. Simpson, Role of aromatase in sex steroid action, J. Mol. Endocrinol. 25 (2000) 149–156. Y. Takema, G. Imokawa, The effects of UVA and UVB irradiation on the viscoelastic properties of hairless mouse skin in vivo, Dermatology 196 (1998) 397–400. J.A. Parrish, S. Zaynoun, R.R. Anderson, Cumulative effects of repeated subthreshold doses of ultraviolet radiation, J. Invest. Dermatol. 76 (1981) 356–358. Y. Takema, M. Hattori, K. Aizawa, The relationship between quantitative changes in collagen and formation of wrinkles on hairless mouse skin after chronic UV irradiation, J. Dermatol. Sci. 12 (1996) 56–63. K. Tsukahara, S. Moriwaki, M. Hotta, T. Fujimura, Y. SugiyamaNakagiri, S. Sugawara, T. Kitahara, Y. Takema, The effect of sunscreen on skin elastase activity induced by ultraviolet-A irradiation, Biol. Pharm. Bull. 28 (2005) 2302–2307. K. Tsukahara, Y. Takema, S. Moriwaki, T. Fujimura, T. Kitahara, G. Imokawa, Age-related alterations of echogenicity in Japanese skin, Dermatology 200 (2000) 303–307. P. Elsner, D. Wilhelm, H.I. Maibach, Mechanical properties of human forearm and vulvar skin, Br. J. Dermatol. 122 (1990) 607–614. P.G. Agache, C. Monneur, J.L. Leveque, J. de Rigal, Mechanical properties and Young’s modulus of human skin in vivo, Arch. Dermatol. Res. 269 (1980) 221–232. H. Miyauchi, T. Horio, Y. Asada, The effect of ultraviolet radiation on the water-reservoir functions of the stratum corneum, Photodermatol. Photoimmunol. Photomed. 9 (1992–1993) 193–197. A. Haratake, Y. Uchida, M. Schmuth, O. Tanno, R. Yasuda, J.H. Epstein, P.M. Elias, W.M. Holleran, UVB-induced alterations in permeability barrier function: roles for epidermal hyperproliferation and thymocytemediated response, J. Invest. Dermatol. 108 (1997) 769–775. E.V. Jensen, E.R. DeSombre, Mechanism of action of the female sex hormones, Annu. Rev. Biochem. 41 (1972) 203–230. L.M. Wahl, R.J. Blandau, R.C. Page, Effect of hormones on collagen metabolism and collagenase activity in the pubic symphysis ligament of the guinea pig, Endocrinology 100 (1977) 571–579. K. Tsukahara, S. Moriwaki, A. Ohuchi, T. Fujimura, Y. Takema, Ovariectomy accelerates photoaging of rat skin, Photochem. Photobiol. 73 (2001) 525–531. C.E. Ihionkhan, K.L. Chambliss, L.L. Gibson, L.D. Hahner, M.E. Mendelsohn, P.W. Shaul, Estrogen causes dynamic alterations in endothelial estrogen receptor expression, Circ. Res. 91 (2002) 814–820. A. Pedram, M. Razandi, M. Aitkenhead, C.C. Hughes, E.R. Levin, Integration of the non-genomic and genomic actions of estrogen. Membrane-initiated signaling by steroid to transcription and cell biology, J. Biol. Chem. 277 (2002) 50768–50775.

90

K. Tsukahara et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 82–90

[22] T. Suzuki, Y. Nakamura, T. Moriya, H. Sasano, Effects of steroid hormones on vascular functions, Microsc. Res. Tech. 60 (2003) 76–84. [23] J. Harvell, I. Hussona-Saeed, H.I. Maibach, Changes in transepidermal water loss and cutaneous blood flow during the menstrual cycle, Contact. Dermatol. 27 (1992) 294–301. [24] N. Charkoudian, J.M. Johnson, Altered reflex control of cutaneous circulation by female sex steroids is independent of prostaglandins, Am. J. Physiol. 276 (1999) H1634–H1640. [25] W. Haenggi, H.R. Linder, M.H. Birkhaeuser, H. Schneider, Microscopic findings of the nail-fold capillaries dependence of menopausal

status and hormone replacement therapy, Maturitas 22 (1995) 37–46. [26] N. Ochiai, K. Yano, S. Denda, N. Akutsu, M. Yanai, Y. Shimatani, E. Takasu, Y. Kano, E. Nozue, T. Uemura, Changes in skin physiological parameters in pre-, peri- and post-menopause and effect of hormone replacement therapy, J. Jpn. Menopause Soc. 8 (2000) 33– 40. [27] K. Hiramoto, S. Takahashi, T. Katada, H. Konishi, M. Sugiura, R. Hayakawa, Sex difference of UV ray sensitivity in mice, Environ. Dermatol. 3 (1996) 321–325.