Y-27632 preserves epidermal integrity in a human skin organ-culture (hSOC) system by regulating AKT and ERK signaling pathways
Abstract
Background: The human skin organ culture (hSOC) developed a century ago has been widely used to study various aspects of human skin development, differentiation, function, disease as well as skin appendages biology, however, maintaining the integrity of epidermal structure in long-term culture, has remained a challenge.
Objectives: Here we tried to establish a culture system using supplemented William’s E medium in the presence of a ROCK inhibitor Y-27632 to maintain epidermal architecture in the long-term hSOC and to investigate the underlying mechanisms.
Methods: Human breast skins, cut into 5 mm × 5 mm pieces, were cultured in supplemented William’s E medium in the presence of 30mM Y-27632. The cultured skin tissues were collected at different time points for analysis of epidermal cell proliferation and differentiation by real time qRT-PCR and immunofluorescence (IF) staining. The keratinocyte suspension assay and in vivo treatment of Y-27632 on mouse were also carried out to study that the regulation of Y-27632 on keratinocyte proliferation and differentiation.
Results: We found Y-27632 not only enhanced both basal cell proliferation and expression of suprabasal cell differentiation markers, but also maintained the balance of keratinocyte proliferation and differentiation through activation of AKT pathways on one hand and inhibition of ERK pathways on the other hand. The AKT inhibitor MK-2206 blocked the epidermal preservation effect of Y-27632, while the MEK/ERK inhibitor U0126 enhanced the preservation of epidermal structure in the hSOC. Conclusions: Y-227632 can maintain skin epidermal integrity through regulation of AKT and ERK activity in the hSOC.
1. Introduction
Human skin organ culture (hSOC), known as an ex vivo model, has been considered to be closer native skin than that produced by in vitro culture. For this reason, hSOC has been widely used for in model studies relating to human skin diseases, pharmacology, physiology and toxicology [1,2]. However, hSOC as a laboratory model is rather infrequently used because it is a big challenge for researchers to preserve the skin intact structure for a sufficient time to perform relevant experiments. The epidermis, which is the outermost layer of skin and acts as a barrier, play a crucial role in maintaining skin homeostasis [3–5]. So, preservation of epidermal structure is crucial for effective skin-environmental interactions. Although long-term viability and histological preservation of the hSOC model has been studied over many decades using various culture media and techniques, little attention has been given to retention of epidermal structure and differentiation [2,6].
Rho-associated protein kinase (ROCK), consisting of two isoforms ROCK1 and ROCK2, originally was discovered to regulate cell shape and migration by acting on the cytoskeleton; however, it actually plays a wide role in modulating many biological processes such as cell survival, proliferation and apoptosis through multiple downstream targets [7,8]. Y-27632, a well-established ROCK inhibitor which blocks, both ROCK1 and ROCK2 activities, has been shown to significantly benefit the culture of a variety of pluripotent stem cell types including human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs) [9–13]. In the skin, Y-27632, has been shown to significantly enhance epidermal stem cell proliferation, and inhibit keratinocyte differentiation, processes which play a significant role in main- taining epidermal architecture [14–17]. Recently, we reported, in BRAF mutant melanoma cells, that Y-27632 could enhance both activation of AKT and ERK signaling pathway [18], which has been shown to play an important role in keratinocyte proliferation and differentiation [19–22]. The present study tests whether Y-27632 in the culture medium could preserve epidermal structure in the hSOC and whether Y-27632 effect on hSOC is through regulation of AKT or ERK signaling pathway. Indeed, we found that the addition of Y-27632 to William’s E medium could significantly prolong the epidermal architecture of hSOC through regulation of the AKT and ERK pathways.
2. Materials and methods
The preparation and culture of hSOC followed the protocol described by Lu et al. [23]. Briefly, human female breast skins (range from 25 to 65 years old) were cut into 5 mm × 5 mm pieces, and were floated onto William’s E medium with the epidermis up at air/liquid interface and the dermis down. The skin pieces of the experiment group were cultured with supplemented William’s E medium in the presence of 30mM Y-27632 (Sigma Chem Co., St. Louis, MO, USA). The cultured skin tissues were collected at different time points as indicated in the figures for the following experiments.
The suspension assay followed previously described [24,25]; hSOC permeability barrier analysis were performed by the Lucifer Yellow dye (L453, Invitrogen) penetration assay followed previ- ously described [26]; Procedures for histological analysis immu- noblotting, immunofluorescence analysis were carried out as previously described [18,27,28]; And all detailed experimental procedures were described in Supplemental Materials and methods.
3. Results
3.1. Y-27632 stabilizes skin epidermal structure of hSOC
Williams E medium was shown to maintain intact skin structure for longer than DMEM medium with 10% FBS [23], therefore Williams E medium was used for all further studies. The skin tissues were cultured in William E medium plus/minus 30mM Y-27632, and harvested at 1 wk, 2, 3 and 4 wks for histological analysis. By light microscopy epidermis of the control skins became much thinner at 2 wks, with nearly one cell layer remaining at 3 wks (blue arrows); and complete loss of epidermis at day 28 (Fig. 1A, red arrows). In contrast, skins maintained in the medium with Y-27632, epidermal structure could be maintained for more than 2 wks, moreover several layers of epidermis remained at 4 wks (Fig. 1A). Interestingly, we didn’t observed a significant difference of dermal structure between control and Y- 27632 treated group even at 3 wks (Suppl. Fig. 1), suggesting that it is a main challenge to maintain the epidermal structure during hSOC, therefore the present study focused on the effect of Y-27632 on preservation of the structure of the epidermis. To further characterize the effect of Y-27632 on the epidermis, the allover thicknesses of the epidermis, which is from cornified layer to the basal layer indicated with double red arrows (Fig. 1B) grown under these two conditions were evaluated. As shown Fig. 1B, the epidermis treated with Y-27632 was significantly thicker than that of the control group. This result was corroborated by quantifying the number of epidermal cells (Fig.1C), counted according to nuclei staining (DAPI, Fig. 1C), showing much more epidermal cells in Y- 27632 treated epidermis at 3 wks. Together, it suggested that Y- 27632 could stabilize the epidermis structure in the hSOC.
3.2. Y-27632 maintains the expression of suprabasal cell marker of the epidermis
A normal highly regulated keratinocyte differentiation program plays a crucial role in maintaining the epidermal integrity, which consists of a undiffereniatied basal cell layer, differentiated suprabasal cell layers and terminal differentiated stratum corneum layer [29]. To analyze the differentiation status of the epidermis, we first studied the expression of keratin 5 (K5), a basal cell marker [30], with immunofluorescence (IF) staining. IF staining of K5 and its quantification data were shown in Fig. 2A and B, we could see that K5 positive cells locate in the lower part of the epidermis in both control and Y-27632 treated groups (red, Fig. 2A). The majority of K5 positive cells were confined to the basal layer of the epidermis, however multiple layers of K5 positive cells were observed at 2 wks in both control and Y-27632 treated skin, but slightly, more K5 positive cells in Y-27632 treated group (Fig. 2B), which p value was 0.047 for the statistical analysis. Next, we analyzed the expression of the suprabasal cell marker keratin 10 (K10) [30], as shown in Fig. 2C-D the basal cells clearly didn’t express K10 (Fig. 2C, red) in both groups. In the control group, K10 positive cells were gradually lost with culture time. At 2 wks, some suprabasal cells showed either lower expression or loss expression of K10 (white arrows, Fig. 2C). At 3 wks, only one single layer of K10 positive cells were observed (white arrowheads, Fig. 2C). In contrast, those skins cultured in Y27632 showed a strong and stable expression of K10 in suprabasal epidermal layers at 2 wks, in addition, several layers of K10 positive cells were still present at 3 wks (Fig. 2C and D), indicating Y-27632 maintaining the expression of suprabasal cell marker K10. Then, we checked expression of the terminal differentiation marker loricrin, a major component of the epidermal cornified envelope(CE) [31]. CE is crucial for skin barrier function, although the loricrin knockout mice presented a mild defect of barrier function mainly due to a compensatory mechanism [32,33]. We could see loricrin expression (green) in the most outer layer of epidermis in both groups (Fig. 2E). In the control group, the integrity of the loricrin-positive layer was disrupted even at 1 week, and some of loricrin positive cells contain nuclei, however the expression level of loricrin was not significantly different between two groups (Fig. 2F). The integrity of loricrin-positive layer was maintained in the Y-27632 treated skin even at 3 wks (Fig. 2F), which was supported by the Lucifer yellow dye penetration assay (Suppl. Fig. 2). It suggested the epidermal barrier function was maintained in hSOC in the presence of Y-27632 until 3 wks. Finally, we detected the expression keratin 6, a hyperproliferating cell marker, previously reported to be suprabasally expressed in the organ cultured skin
[34]. Agreed to the previous report, K6 expression was detected in suprabasal layers of both control and Y-27632 treated epidermis (Suppl. Fig. 3) at 1 wk. And at 3 wks, K6 expression was nearly lost in the control epidermis, but maintained in the Y-27632 treated group. This data further supported Y-27632 can maintain the suprabasal layer in the hSOC. Taken together, we could conclude that Y-27632 preserves expression of keratinocyte differentiation markers, such as K5, K10 and loricrin, presented in the right place of epidermis during the organ-culture, especially, it enhances maintaining suprabasal layer, resulting in maintenance of the epidermal integrity in the hSOC.
3.3. Y-27632 enhances keratinocyte proliferation and expression of suprabasal differentiation markers in the skin organ culture
In order to understand how Y-27632 treatment maintain the integrity of epidermis during hSOC, we analyzed epidermal cell proliferation, apoptosis and differentiation in the early time points of culture. Using the keratinocyte proliferation marker Ki67, we found proliferating cells (green, red arrows, Fig. 3A) located in the epidermal basal layer of both control and Y-27632 treated group; more Ki67 positive cells were observed in Y-27632 group (red arrows, Fig. 3A). Quantification analysis showed that the percentage of proliferating cells in the control epidermis was reduced from 24 h to 72 h of culture, while that of the Y-27632 treated group showed an increase (Fig. 3B). The increased proliferation of epidermal cells in skin treated with Y-27632 was further confirmed by EdU incorporation assay (Fig. 3C and D), more EdU incorporated cells (green, red arrows, Fig. 3C) were found in Y- 27632 treated group. Then we tested whether Y-27632 could reduce epidermal cell apoptosis to maintain the epidermal layers, the apoptotic cells were detected by cleaved caspase 3 staining (Fig. 3E), and no significant difference for the percentage of apoptotic cells between control and Y-27632 treated group, indicating Y-27632 didn’t reduce the epidermal cell apoptosis. Next, we analyzed the effect of epidermal differentiation by Y- 27632 treatment, we carried out IF analysis of differentiation markers K5, K10 and loricrin at 72 h, and quantification analysis of IF staining in Fig. 3H, showing stronger staining of K10 in Y-27632 treated skin; there was no clear difference for K5 and loricrin expression level between control and Y-27632 treated skin. To further confirm Y-27632 enhance the expression of suprabasal markers such as K10, the real time qRT-PCR analysis for the expression of differentiation markers at 24 h, 48 h and 72 h after treatment of Y-27632 were performed (Suppl. Fig. 4). The PCR results showed increased expression of both suprabasal markers keratin 1 (K1) [30] and K10 in Y-27632 treated skin at all time points (Suppl. Fig. 4). There was an increased expression of terminal differentiation markers loricrin and filaggrin, a molecule linked to the cornified envelope [30,35], in Y-27632 treated skin at 24 h, but no clear difference at 48 h and 72 h (Suppl. Fig. 4), which supporting the staining result in Fig. 3G. Taking together, these data suggested that Y-27632 increases keratinocyte proliferation and expression of differentiation markers, K1 and K10, during the organ-culture.
3.4. Y-27632 promotes K1 and K10 expression of keratinocytes both in vitro suspension culture and in vivo
To further validate that Y-27632 treatment could enhance the expression of superabasal cell markers K1 and K10, two approaches were carried out. First, in vitro suspension assay, which is a frequently used method to induce differentiation besides modulating extracellular Ca2+ [24], was performed to test the effect Y-27632 on proliferation and differentiation on suspended cells. The suspended keratinocytes were collected for analysis at different time points by qRT-PCR analysis. As shown in Fig. 4A, Y-27632 increased Ki67 expression, and in Fig. 4B and C Y- 27632 induced K1 and K10. These results agreed with the above observations of hSOC; however, we found that Y-27632 signifi- cantly decreased the expression of the terminal differentiation markers Loricrin and filaggrin in the suspended cells (Fig. 4D and E). The PCR results were confirmed by the western-blot analysis of keratinocytes lysed at different time points (Fig. 4F). Secondly, we repeated the study on in vivo skin preparations to validate the Y- 27632 effects on expression of differentiation markers. The mouse skins treated with Y-27632 were collected for IF analysis, Fig. 4G and H shows stronger staining for K10 expression in the Y-27632 treated mouse skin, and the loricrin expression was not decreased by Y-27632 compared to the control mouse. The in vivo result agrees that of the organ-culture, suggesting hSOC is likely closer to in vivo system, compare to the suspension assay. Taken together, these data further supported that Y-27632 can enhance keratino- cyte K1 and K10 expression in the skin both in vitro and in vivo.
3.5. Y-27632 activates AKT, but inhibits ERK activity in keratinocytes
To understand the Y-27632 mechanism affecting hSOC kera- tinocyte proliferation and differentiation, we analyzed the AKT and ERK pathways, which have been reported to play crucial roles in regulating keratinocyte proliferation and differentiation. To mimic the condition of hSOC, keratinocytes were cultured in suspension treated with or without Y-27632 and collected at different time points as indicated in Fig. 5A and B and then analyzed by western blot analysis for the phosphorylation forms of AKT and ERK (Fig. 5C). As shown in Fig. 5A after 6 h of suspension, the activity of AKT was significantly increased in the Y-27632 treated keratino- cytes compared to the control cells (Fig. 5A). This higher expression of activated AKT persisted up to 72 h (Fig. 5B). In contrast, the activity of ERK decreased in Y-27632 treated cells starting at 1 h after suspension. It is notable that the difference of phospho-ERK expression between control and Y-27632 treated group diminished with time so that it was not apparent at 72 h (Fig. 5B). In the absence of Y-27632 (control), ATK activity decreased while ERK activity increased with culture time (Fig. 5C). The effect of Y-27632 on ERK activation was further confirmed by western-blot analysis of its downstream target p90RSK, and the pattern of p90RSK activation induced by Y-27632 was similar to that of ERK (Suppl. Fig. 5). Moreover, the increased AKT activity and inhibition of ERK by Y-27632 were further validated by staining of phospho-AKT and phospho-ERK in skin derived from hSOC at 1 wk (Fig. 5D, F) and corresponding quantification analysis (Fig. 5E, G).
3.6. Y-27632 maintains hSOC skin structure by regulating AKT and ERK activity
In order to verify whether Y-27632 regulates keratinocyte proliferation and differentiation through AKT and ERK pathway, we cultured keratinocytes in suspension with/without Y-27632 in the presence of MK-2206, which inhibits AKT activity or U0126, MEK inhibitor to block ERK pathway. At 48 h, cells were lysed for qRT- PCR analysis of proliferation and differentiation markers. Fig. 6A showed that the increased Ki67 expression induced by Y-27632 was blocked by MK-2206, but not by U0126, suggesting the increased proliferation by Y-27632 likely occurred through activation of AKT. The increased K1/K10 expression induced by Y-27632 could be blocked by both AKT and ERK inhibitors (Fig. 6B and C). The AKT inhibitor didn’t significantly decrease the expression of terminal differentiation markers and couldn’t rescue the inhibition of terminal differentiation induced by Y-27632, however ERK inhibitor did block induced differentiation of cells in suspension, regardless of the presence of Y-27632 (Fig. 6D–F). Then, we tested whether Y-27632 maintains hSOC epidermal structure by regulating AKT and ERK activation. Fig. 6G shows that at 1 wk the integrity of epidermis was maintained in all groups (Fig. 6G). At 3 wks, the epidermis was almost lost in the control group, which is similar to as shown in Fig. 1A, and most of epidermis in MK-2206 treated group was detached from the dermis (black arrows, Fig. 6G). Although epidermal structure in the U0126 treated group was better than control and MK-2206 treated group, it was still thinner than that of skin treated with Y-27632, suggested inhibition ERK alone might be not sufficient to maintain the integrity of epidermal structure. These results were corrobo- rated by quantification analysis of epidermal thickness and number of viable epidermal cells (Fig. 6H). Taking together, the data suggest that Y-27632 maintains hSOC epidermal structure by regulating both AKT and ERK activation.
4. Discussions
An ideal culture system, should not only preserve epidermal viability, but it should also preserve proliferation and normal differentiation of epidermal cells [2]. In early organ-culture systems, serum was considered an essential component necessary to maintain the skin viability and integrity, but as serum is chemically undefined and often variable presents difficulties for reproducible skin culture studies [2]. Zeltinger et al. tested 11 types of culture media to grow fetal skin tissues for 3–4 wks and found that serum free medium DMEM/F12 was optimal, suggesting that defined serum free media was able to maintain long-term skin viability and integrity in hSOC [36]. Lu et al. applied supplemented serum-free William’s E medium to culture adult human scalp skin and observed active hair shaft growth until day 16; however there was progressive thinning and hyperkeratosis of the interfollicular epidermis after day 5(23). By using the same medium, Kleszc- zynski and Fisher showed that epidermal proliferation and structure of abdominal skin could be maintained for only 48 h [37], but that short-term cultures could be still be used for meaningful study of epidermal function [38]. Recently, Buckingham et al. used supplemented MEM to culture foreskins for 28 days and studied the autophagy response of human neonatal skin after infection with varicell-zoster virus (VZV); unfortunately the epidermal structure was not analyzed in this report [39]. Nevertheless, these studies did not address the issue of epidermal structure-function preservation. Here we carefully characterized the epidermal structure by both detailed histological analysis and immunofluo- rescence staining of epidermal differentiation status at different culture time points and discovered that the incorporation of the Rock inhibitor Y-27632 into the supplemented William’s E medium significantly prolongs the preservation of skin epidermis architec- ture up to at least 3 wks and maintained keratinocyte proliferation and differentiation.
The normal keratinocyte proliferation and differentiation is highly regulated, plays a crucial role for maintaining epidermal integrity [29,40]. By analysis of differentiation and proliferation status, we found that Y-27632 preserves basal layer proliferation and suprabasal layer expression of K1 and K10 (Figs. 2C and D, 3, 4) during the culture. Interestingly, Y-27632 significantly inhibits the expression of loricrin and filaggrin in the suspension assay (Fig. 4D–F), but not in the hSOC (Figs. 3F and G, 4G and H). in vivo treatment of Y-27632 did not block keratinocyte terminal differentiation (in mouse), thus supporting the findings with hSOC. Importantly, our findings fully agreed with findings of previous reports [14,15]. It has been reported that Y-27632 could block terminal differentiation of loricrin mostly through inhibition of ROCK2 but increase early differentiation markers of K1 and K10 through inhibition of ROCK1 in the in vitro suspension assay [15]. However, it has been also reported that inhibition of ROCK actually induced terminal differentiation [41]. Therefore, the difference between in vitro suspension assay and the hSOC might be due to their different physiological microenvironments.
How Y-27632 regulates keratinocyte proliferation and differentiation has not been fully documented so far. We recently found Y-27632 could promote BRAF mutant melanocyte growth and migration through regulation of PI3K/AKT and RAF/ERK pathways [18]; both pathways have been shown to play a crucial role in regulating keratinocyte proliferation and differentiation [19,20,42–46]. Peng et al. reported that double knockout of AKT1 and ATK2 in mouse resulted in much thinner of skin layers, especially of the suprabasal (spinous) layer, compared to the control little mate, and they found that decreased proliferation of basal cells, and significantly reduced expression of differentiation markers such as K10 and filaggrin [45]. The study from Calautti et al. demonstrated that the PI3K/AKT pathway was activated during mouse keratinocyte differentiation, that active expression of AKT could promoted expression of differentiation markers K1 and loricrin, and that this effect could be blocked by AKT inhibitor MK-2206 [19]. Here we found Y-27632 could significantly increase AKT activity both in suspension and in the hSOC, and the AKT inhibitor MK-2206 could completely block both increased prolif- eration and differentiation induced by Y-27632 (Fig. 6). Therefore, we think that Y-27632 enhanced keratinocyte proliferation and increased the expression of K1/K10 differentiation in the hSOC by way of AKT activation. ROCK was shown to activate PTEN, which negatively regulate activity of PI3K [47], suggesting Y-27632 enhances AKT activation in keratinocytes probably through regulation of PTEN activation.
Besides the PI3K/AKT pathway, RAF-ERK signaling pathway is another mechanism for controlling cell survival, differentiation and proliferation in response to extracellular cues [21,44,48]. Our results showed that Y-27632 could significantly inhibit the activation of ERK in keratinocytes. It has been reported that RAF-MEK-ERK pathway could negatively regulate epidermal differentiation, and the activation of this pathway in epidermis will increase keratinocyte proliferation and block its differentia- tion [43,44], and that indicating downregulation of RAF-ERK activation would promote keratinocyte differentiation. Therefore, inhibition of ERK activity by Y-27632 could contributed to enhance keratinocyte differentiation. However, when we used U0126, a MEK inhibitor to block the ERK signaling pathway, we found that it actually blocked keratinocyte differentiation, and decreased the expression of terminal differentiation markers, such as loricrin, filaggrin, even in the presence of Y-27632. Recent publications showed that ERK inhibitors could block keratinocyte differentia- tion, just as we found here [22,49]. Therefore, the data suggest that Y-27632 possibly blocks keratinocyte differentiation through decreasing RAF-ERK activation, which to some extent, prevents a potential hyper-differentiation induced by the sustained ATK activation with the continuous treatment of Y-27632. Interestingly, the addition of U0126 into William’s E medium prolongs the preservation of skin epidermal structure in the hSOC, which partially mimics the result from Y-27632 treatment (Fig. 6). It indicates that downregulation of ERK activity by Y-27632 is crucial for preserving epidermal integrity in the hSOC.
In summary, we report here that Rock inhibitor Y-27632 enhances the preservation of epidermal integrity in hSOC, very likely by increasing AKT and decreased ERK activities. A stable long-term hSOC system will definitely widen its future applica- tions.