Lysates of a Probiotic, Lactobacillus rhamnosus, Can Improve Skin Barrier Function in a Reconstructed Human Epidermis Model

The main function of the skin is to protect the body from the external environment. The barrier function of the skin is mainly provided by the stratum corneum, which consists of corneocytes bound with the corneodesmosomes and lamellar lipids. Skin barrier proteins like loricrin and filaggrin also contribute to the skin barrier function. In various skin diseases, skin barrier dysfunction is a common symptom, and skin irritants like detergents or surfactants could also perturb skin barrier function. Many efforts have been made to develop strategies to improve skin barrier function. Here, we investigated whether the microfluidized lysates of Lactobacillus rhamnosus (LR), one of the most widely used probiotic species for various health benefits, may improve the skin barrier function in a reconstructed human epidermis, Keraskin™. Application of LR lysate on Keraskin™ increased the expression of tight junction proteins; claudin 1 and occludin as determined by immunofluorescence analysis, and skin barrier proteins; loricrin and filaggrin as determined by immunohistochemistry and immunofluorescence analysis and qPCR. Also, the cytotoxicity of a skin irritant, sodium lauryl sulfate (SLS), was alleviated by the pretreatment of LR lysate. The skin barrier protective effects of LR lysate could be further demonstrated by the attenuation of SLS-enhanced dye-penetration. LR lysate also attenuated the destruction of desmosomes after SLS treatment. Collectively, we demonstrated that LR lysate has protective effects on the skin barrier, which could expand the utility of probiotics to skin-moisturization ingredients.


Introduction
The skin is the largest organ of the body, which protects our body from the harmful environment and helps to regulate our body temperature. The skin is composed of three layers, epidermis, dermis, and the hypodermis. The outermost layer of the skin, the epidermis, is a multi-layered epithelial tissue, and the barrier function of the skin is mainly provided by the stratum corneum. The stratum corneum serves as a selective permeability barrier against infectious or toxic substances [1]. The stratum corneum is composed of terminally differentiated keratinocytes called corneocytes, that are attached to each other by corneodesmosomes and lamellar lipids. The structural organization of this barrier is also referred as a 'brick and mortar' structure with the keratin microfibrils, filaggrin and cornified envelopes forming the bricks and the lipids forming the mortar to seal together the cornified envelopes [2]. Also, 3  To further analyze the effects of LR on the epidermis, the treated tissues underwent immunofluorescence (IF) staining with antibodies against target junction proteins, claudin1, and occludin. IF images show that both tight junction molecules were increased in LR-treated Keraskin TM (Figure 2). To further analyze the effects of LR on epidermal differentiation, the treated tissues underwent immunohistochemistry with antibodies against cytokeratin 5 (K5), 1 (K1), 10 (K10), loricrin (LOR) and filaggrin (FLG). As shown in Figure 3a and in the scoring of intensity (1 to 3) for each layer of the epidermis (Table 1), LR-treated tissues showed that K5 and K1 advanced into the upper layer of the epidermis; granular layer (GL) and cornified layer (CL), and the intensity increased while control tissues showed K5 and K1 expression limited within the basal layer. In the case of loricrin and filaggrin, there was no significant difference in the localization, but the intensity increased. There was no evident difference in the expression of K10. This pattern was further confirmed with immunofluorescence analysis (Figure 3b). To further analyze the effects of LR on the epidermis, the treated tissues underwent immunofluorescence (IF) staining with antibodies against target junction proteins, claudin1, and occludin. IF images show that both tight junction molecules were increased in LR-treated Keraskin TM ( Figure 2). To further analyze the effects of LR on the epidermis, the treated tissues underwent immunofluorescence (IF) staining with antibodies against target junction proteins, claudin1, and occludin. IF images show that both tight junction molecules were increased in LR-treated Keraskin TM (Figure 2). To further analyze the effects of LR on epidermal differentiation, the treated tissues underwent immunohistochemistry with antibodies against cytokeratin 5 (K5), 1 (K1), 10 (K10), loricrin (LOR) and filaggrin (FLG). As shown in Figure 3a and in the scoring of intensity (1 to 3) for each layer of the epidermis (Table 1), LR-treated tissues showed that K5 and K1 advanced into the upper layer of the epidermis; granular layer (GL) and cornified layer (CL), and the intensity increased while control tissues showed K5 and K1 expression limited within the basal layer. In the case of loricrin and filaggrin, there was no significant difference in the localization, but the intensity increased. There was no evident difference in the expression of K10. This pattern was further confirmed with immunofluorescence analysis (Figure 3b). To further analyze the effects of LR on epidermal differentiation, the treated tissues underwent immunohistochemistry with antibodies against cytokeratin 5 (K5), 1 (K1), 10 (K10), loricrin (LOR) and filaggrin (FLG). As shown in Figure 3a and in the scoring of intensity (1 to 3) for each layer of the epidermis (Table 1), LR-treated tissues showed that K5 and K1 advanced into the upper layer of the epidermis; granular layer (GL) and cornified layer (CL), and the intensity increased while control tissues showed K5 and K1 expression limited within the basal layer. In the case of loricrin and filaggrin, there was no significant difference in the localization, but the intensity increased. There was no evident difference in the expression of K10. This pattern was further confirmed with immunofluorescence analysis (Figure 3b

LR Lysate Increases the Expression of Skin Barrier Proteins, Loricrin and Filaggrin in Keraskin TM
To confirm IHC and IF data ( Figure 3) and investigate whether LR could promote epidermal differentiation indeed, mRNA levels of loricrin and filaggrin that are major proteins for the formation of the epidermal barrier were evaluated using qPCR. LR lysate was topically applied to Keraskin TM for 2 days and mRNA was isolated as described in the Methods section and qPCR was conducted. As a result, LR-treatment significantly increased the expression of loricrin and filaggrin (by about 60% and 40%, respectively, Figure 4), thus confirming the IHC results.

LR Lysate Increases the Expression of Skin Barrier Proteins, Loricrin and Filaggrin in Keraskin TM
To confirm IHC and IF data ( Figure 3) and investigate whether LR could promote epidermal differentiation indeed, mRNA levels of loricrin and filaggrin that are major proteins for the formation of the epidermal barrier were evaluated using qPCR. LR lysate was topically applied to Keraskin TM for 2 days and mRNA was isolated as described in the Methods section and qPCR was conducted. As a result, LR-treatment significantly increased the expression of loricrin and filaggrin (by about 60% and 40%, respectively, Figure 4), thus confirming the IHC results.

Topical Treatment of LR Lysate Attenuated Irritant-Induced Cytotoxicity in Keraskin™
Skin barrier protective effects can be examined by determining the protection of irritant-induced cell death [21]. LR lysate was applied to Keraskin™ and an irritant, SLS was treated at various concentrations. As a result, treatment of LR significantly attenuated SLS-induced cell death in Keraskin™ as measured by WST-1 assay ( Figure 5a). To visualize the protection effects of LR, Keraskin™ tissues were stained with H&E ( Figure 5b). Histologic images also showed that LR treated epidermal tissues had more intact stratum corneum than the control both in the presence and absence of SLS, which was in line with Figure 1b. Especially, while SLS treatment resulted in the almost complete disappearance of stratum corneum, LR treated tissues showed stratum corneum remaining.
(a) (b) Figure 5. Attenuation of irritant-induced cytotoxicity by LR in Keraskin TM . Tissues were pretreated with LR for every other day 4 times and SLS was treated for 10 min. (a) Tissue viability was determined by the WST-1 assay. Results are expressed by the fold of control viability as mean ± SE of independent more than three tests and analyzed by two-way ANOVA. * p < 0.05 compared with PBStreated control. (b) H&E stained tissues after with PBS or SLS treatment were measured by the light microscopy.

Topical Treatment of LR Lysate Reduced the Skin Penetration of Rhodamine B
We further studied the skin barrier protection effect of LR through a skin penetration study. Rhodamine B, a fluorescent dye, was used to evaluate the permeability of the skin. Keraskin™ was treated with LR for 24 h and 0.02% rhodamine B was treated for 2 h. After fixation with OCT compound and cryosection, the permeability of rhodamine B was evaluated with a fluorescence

Topical Treatment of LR Lysate Attenuated Irritant-Induced Cytotoxicity in Keraskin™
Skin barrier protective effects can be examined by determining the protection of irritant-induced cell death [21]. LR lysate was applied to Keraskin™ and an irritant, SLS was treated at various concentrations. As a result, treatment of LR significantly attenuated SLS-induced cell death in Keraskin™ as measured by WST-1 assay (Figure 5a). To visualize the protection effects of LR, Keraskin™ tissues were stained with H&E ( Figure 5b). Histologic images also showed that LR treated epidermal tissues had more intact stratum corneum than the control both in the presence and absence of SLS, which was in line with Figure 1b. Especially, while SLS treatment resulted in the almost complete disappearance of stratum corneum, LR treated tissues showed stratum corneum remaining.

Topical Treatment of LR Lysate Attenuated Irritant-Induced Cytotoxicity in Keraskin™
Skin barrier protective effects can be examined by determining the protection of irritant-induced cell death [21]. LR lysate was applied to Keraskin™ and an irritant, SLS was treated at various concentrations. As a result, treatment of LR significantly attenuated SLS-induced cell death in Keraskin™ as measured by WST-1 assay (Figure 5a). To visualize the protection effects of LR, Keraskin™ tissues were stained with H&E ( Figure 5b). Histologic images also showed that LR treated epidermal tissues had more intact stratum corneum than the control both in the presence and absence of SLS, which was in line with Figure 1b. Especially, while SLS treatment resulted in the almost complete disappearance of stratum corneum, LR treated tissues showed stratum corneum remaining.

Topical Treatment of LR Lysate Reduced the Skin Penetration of Rhodamine B
We further studied the skin barrier protection effect of LR through a skin penetration study. Rhodamine B, a fluorescent dye, was used to evaluate the permeability of the skin. Keraskin™ was treated with LR for 24 h and 0.02% rhodamine B was treated for 2 h. After fixation with OCT compound and cryosection, the permeability of rhodamine B was evaluated with a fluorescence Figure 5. Attenuation of irritant-induced cytotoxicity by LR in Keraskin TM . Tissues were pretreated with LR for every other day 4 times and SLS was treated for 10 min. (a) Tissue viability was determined by the WST-1 assay. Results are expressed by the fold of control viability as mean ± SE of independent more than three tests and analyzed by two-way ANOVA. * p < 0.05 compared with PBS-treated control. (b) H&E stained tissues after with PBS or SLS treatment were measured by the light microscopy.

Topical Treatment of LR Lysate Reduced the Skin Penetration of Rhodamine B
We further studied the skin barrier protection effect of LR through a skin penetration study. Rhodamine B, a fluorescent dye, was used to evaluate the permeability of the skin. Keraskin™ was treated with LR for 24 h and 0.02% rhodamine B was treated for 2 h. After fixation with OCT compound and cryosection, the permeability of rhodamine B was evaluated with a fluorescence microscope. As shown in Figure 6, fluorescence images confirmed that the permeability of rhodamine B (red) was reduced by treatment of LR both in the presence and absence of SLS. microscope. As shown in Figure 6, fluorescence images confirmed that the permeability of rhodamine B (red) was reduced by treatment of LR both in the presence and absence of SLS.

LR Lysate Protects the Desmosome Degradation in Keraskin™
Desmosomes are intercellular junctions that provide mechanical strength to tissues. They are abundant in epithelial tissues. As desmosomes are one of the essential factors of the skin barrier function, we studied the effect of LR on desmosomes by the transmission electron microscopy (TEM) (Figure 7). When Keraskin™ was disrupted by SLS, the number of desmosomes was decreased in both control and LR treated groups but LR -treated Keraskin™ showed that desmosomes were damaged less by SLS and more preserved than the control groups.'

LR Lysate Protects the Desmosome Degradation in Keraskin™
Desmosomes are intercellular junctions that provide mechanical strength to tissues. They are abundant in epithelial tissues. As desmosomes are one of the essential factors of the skin barrier function, we studied the effect of LR on desmosomes by the transmission electron microscopy (TEM) (Figure 7). When Keraskin™ was disrupted by SLS, the number of desmosomes was decreased in both control and LR treated groups but LR -treated Keraskin™ showed that desmosomes were damaged less by SLS and more preserved than the control groups.

Discussion
Here, we demonstrated that the lysates of Lactobacillus rhamnosus (LR), one of the most widely used probiotic strains for various health benefits, can improve the skin barrier function using a reconstructed human epidermis model, Keraskin™. Application of LR lysate on Keraskin™ increased

Discussion
Here, we demonstrated that the lysates of Lactobacillus rhamnosus (LR), one of the most widely used probiotic strains for various health benefits, can improve the skin barrier function using a reconstructed human epidermis model, Keraskin™. Application of LR lysate on Keraskin™ increased the expression of target junction proteins, claudin1 and occludin as seen in immunofluorescence images and the expression of skin barrier proteins, loricrin and filaggrin at both protein and mRNA levels. Moreover, the beneficial effects of LR lysate on the skin barrier function could be demonstrated by the attenuation of the SLS-induced cytotoxicity and skin permeability in Keraskin™. Application of LR lysate also attenuated the destruction of desmosomes by SLS.
Keratins form the intermediate filaments which become crosslinked as a cytoskeleton for the cornified envelope, contributing to the stability of the epidermal barrier. The mutations of the keratins cause incomplete intermediate filaments and disturbed barrier function [22]. One of the late epidermal differentiation proteins, filaggrin aggregates the keratin filaments. Many studies demonstrated that the loss of function in filaggrin causes the impaired formation of the epidermis and affect the skin barrier function [23,24]. Loricrin is a major component of the cornified envelope and binds with filaggrin for the formation of the cornified envelope and plays a key role in the skin barrier function [22]. Immunohistochemical images showed that the intensity of cytokeratins 5 and 1 increased and found at the upper layer of the epidermis of LR-treated tissues and the intensity of loricrin and filaggrin also increased, reflecting that LR may promote epidermal differentiation. This was confirmed through qPCR analysis of loricrin and filaggrin. This indicates that the promotion of epidermal differentiation by LR lysate may have increased barrier function through strengthening the stratum corneum, which could be observed with H&E stained tissues.
SLS is an anionic surfactant widely used in various skincare products like soaps and shampoos. SLS causes skin damages through disrupting stratum corneum integrity. We could also observe that the stratum corneum of Keraskin TM was severely damaged by SLS. In addition, it is known that SLS alters the expression pattern of keratinocyte differentiation markers [25]. We demonstrated that LR treatment causes the increased expression of skin barrier proteins at both protein and mRNA levels. So, it is highly likely that LR could enhance the stability and impermeability of the stratum corneum in the presence of SLS though strengthening the stratum corneum, as shown in the histology and permeation study.
Akt activation in keratinocytes promotes epidermal cell differentiation with the increased expression levels of loricrin and filaggrin [26]. Akt activity is also essential for the barrier, especially in the formation of the cornified envelope during the late and terminal differentiation [27]. Akt pathway is generally activated by several extracellular signals such as growth factor, cytokines through phosphoinositide 3-kinase (PI3K)-dependent signaling pathway. Especially, Toll-like receptor 2 (TLR2) is the pattern recognition receptor expressed beneath the stratum corneum and serves to recognize the presence of bacteria. Recently, TLR2 activation resulted in the activation of the PI3K/Akt pathway and has been shown to enhance epithelial barrier function. It is demonstrated that L. rhamnosus promotes TLR2/Akt activation in response to bacterial infection, thus protecting cells by maintaining the epithelial barrier and promoting intestinal epithelial cell activation against an enterotoxigenic Escherichia coli [16]. In addition, it has been known that bacterial peptidoglycan-induced activation of TLR2 enhances expression of tight junction proteins such as claudins, occludin and ZO proteins thus resulting in enhanced barrier function in keratinocytes [28]. However, it is also reported that LR-augmented tight junction function is not mediated by TLR2 in human keratinocytes [18]. It is thought that topical LR treatment on the skin is derived by different mechanisms from the gut and also application of whole lysates could undergo multiple signaling pathways for barrier protection effects. Currently, further studies are necessary for how LR enhances the skin barrier function.
We examined whether the lysates of LR can have skin barrier protective effects. It would be interesting to examine whether the application of live L. rhamnosus can manifest skin barrier protective effects. However, in cosmetics, products must not be contaminated with microorganisms including probiotic species. The FDA regulates that acceptable limits for total (not pathogenic) microorganisms in cosmetics are 500 colony forming units (cfu) per gram in eye-area products and 1000 cfu/g for other area products [29]. So at this moment, the use of live probiotic strains in cosmetic products must overcome this hurdle while other biotherapeutic approaches beyond cosmetics are on-going with live probiotics [30]. If this hurdle could be mitigated by the amendment of cosmetic regulation, new cosmetics for skin moisturization may be possible.
In conclusion, our results support LR has a beneficial effect on the maintaining of skin barrier function. We demonstrated novel skin barrier protection effects of employing a reconstructed human epidermis model. Further studies will be required to elucidate the mechanism of the skin barrier protection effect of LR. With additional researches, LR could be used for the skincare products as the topical probiotics.

A Reconstructed Human Epidermis Model (Keraskin™)
A reconstructed human epidermis model (Keraskin™) [31] and Keraskin™ culture media were purchased from Biosolution Co., Ltd. (Seoul, Korea). Keraskin™ was placed on a six-well plate filled with 0.9 mL Keraskin™ culture media per well, and pre-incubated for 20-24 h at 37 • C in a humidified atmosphere containing 5% CO 2 . After pre-incubation, tissues were treated with 40 µL of various concentrations of LR diluted in PBS. After 48 h treatment, tissues were rinsed with PBS. LR was treated four or eight times, and control tissues were treated with PBS and followed the same schedule. Culture media was changed every 48 h. After four or eight times of treatment, tissues were treated with various concentrations of SLS diluted in PBS for 10 min and then rinsed with PBS to perturb skin condition.

WST-1 Assay
Tissue viability was determined by WST-1 assay which gives less damage to skin tissues [32]. After incubation, WST-1 diluted in sterile PBS was treated 300 uL for each tissue and incubated for 3 h at 37 • C in a 5% CO 2 incubator. Then 200 µL of solutions were transferred into a 96-well plate and absorbance was determined by microplate spectrophotometer at 450 nm (BioTek Instruments, Inc., Winooski, VT, USA).

Histological Analysis
For the histological examination, all samples were cut in 10 mm-width and fixed in 4% phosphate-buffered formalin (PFA) for 24 h, as described previously [33]. Fixed samples were paraffin-embedded and cut into 5-µm section using microtome (Leica, RM2235), followed by hematoxylin-eosin staining. Briefly, the sections were stained with 0.1% Mayer's hematoxylin for 10 min and 0.5% eosin in 95% EtOH. After staining with hematoxylin and eosin, the washing steps were immediately and sequentially proceeded as follows: dip in distilled H 2 O until eosin stops streaking, dip in 50% EtOH for 10 times, dip in 70% EtOH for 10 times, incubated in 95% EtOH for 30 sec and 100% EtOH for 1 min. Then, samples were covered with a mounting solution (Thermo Scientific, 6769007) and examined under the light microscope (OLYMPUS, BX43).

Immunohistochemistry (IHC) and Immunofluorescence Staining (IF)
For immunohistochemistry, skin samples were cut in cut into 5-µm section and were sequentially proceeded rehydration steps with descending graded series of ethanol. Next, pH 6.0 antigen retrieval (DAKO, S1699, Santa Clara, CA, USA) was conducted using a high-pressure cooker for 15 min, followed the cooling phase over 1 h until the solution was fully transparent. After two washes in PBS, sections were incubated in 3% H 2 O 2 for 30 min for blocking endogenous peroxidase. Another two washes in PBS, sections were incubated with protein block (DAKO, X0909, Santa Clara, CA, USA) for 1-2 h at room temperature in a humidity-controlled chamber. Primary antibodies were incubated overnight at 4 • C. Commercial primary antibodies were purchased and diluted as described below: Anti-keratin1 (ab185628, 1:1k) anti-keratin5 (ab52635, 1:500), anti-keratin10 (ab76318, 1:3k), anti-loricrin (ab24722, 1:500) anti-filaggrin (NBP1-87528, 1:250). After three washes in PBS, sections were incubated in HRP-labeled anti-rabbit antibody (DAKO, K4003, Santa Clara, CA, USA) for 15 min at room temperature. For the development of HRP-labeled antibody on section, DAB (DAKO, K3468, Santa Clara, CA, USA) was diluted and put it on each section for the identical time. Mayer s hematoxylin (DAKO, S3309, Santa Clara, CA, USA) was used for counterstaining.

RNA Preparation and Quantitative Real-Time PCR (qRT-PCR) for the Determination of mRNA Expression
Total RNA was isolated from Keraskin TM by RNA extraction kit according to manufacturer s guidelines (Qiagen, Valencia, CA, USA) and a previous report [34]. Then isolated RNA was quantified by spectrophotometer (NanoDrop spectrophotometer; Thermo Scientific, Fremont, CA, USA). Total RNA was reverse transcribed into cDNA using RT master premix with oligo dT (Elpis Biotech Inc., Seoul, Korea) then analyzed by qRT-PCR using an Applied Biosystems 7300 real-time PCR System (Applied Biosystems, Waltham, MA, USA). The sequences of PCR primers were: for loricrin 5 -CCAGACCCAGCAGAAGCAG-3 (forward) and 5 -TGCCCCTGGAAAACACCTC-3 (reverse), filaggrin 5 -GAGGGCACTGAAAGGCAAAA-3 (forward) and 5 -CTTCCGTGCT GAGAGTGTCT-3 (reverse), β-actin 5 -GCAAAGACCTGTACGCCAAC-3 (forward) and 5 -GATCTT CATTGTGCTGGGTGC-3 (reverse). Primers for human loricrin, filaggrin and β-actin were purchased from Cosmo Genetech (Seoul, Korea). Cycling parameters were 50 • C for 2 min, 95 • C for 10 min, 40 cycles of 95 • C for 15 s, and 58 • C for 1 min. For the comparisons, mRNA levels of target genes were normalized to the corresponding β-actin levels in the Keraskin TM and expressed as target gene normalized to β-actin. Relative quantification was performed using the comparative ∆∆Ct method according to the manufacturer s instructions.

Conflicts of Interest:
The authors declare no conflict of interest.