Licochalcone A Protects the Blood–Milk Barrier Integrity and Relieves the Inflammatory Response in LPS-Induced Mastitis

Background/Aims: Mastitis is an acute clinical inflammatory response. The occurrence and development of mastitis seriously disturb women's physical and mental health. Licochalcone A, a phenolic compound in Glycyrrhiza uralensis, has anti-inflammatory properties. Here, we examined the effect of licochalcone A on blood-milk barrier and inflammatory response in LPS-induced mice mastitis. Methods: In vivo, we firstly established mice models of mastitis by canal injection of LPS to mammary gland, and then detected the effect of licochalcone A on pathological indexes, inflammatory responses and blood-milk barrier in this model. In vivo, Mouse mammary epithelial cells (mMECs) were treated with licochalcone A prior to the incubation of LPS, and then the inflammatory responses, tight junction which is the basic structure of blood-milk barrier were analyzed. Last, we elucidated the anti-inflammatory mechanism by examining the activation of mitogen-activated protein kinase (MAPK) and AKT/NF-κB signaling pathways in vivo and in vitro. Result: The in vivo results showed that licochalcone A significantly decreased the histopathological impairment and the inflammatory responses, and improved integrity of blood-milk barrier. The in vitro results demonstrated that licochalcone A inhibited LPS-induced inflammatory responses and increase the protein levels of ZO-1, occludin, and claudin3 in mMECs. The in vivo and in vitro mechanistic study found that the anti-inflammatory effect of licochalcone A in LPS-induced mice mastitis was mediated by MAPK and AKT/NF-κB signaling pathways. Conclusions and Implications: Our experiments collectively indicate that licochalcone A protected against LPS-induced mice mastitis via improving the blood–milk barrier integrity and inhibits the inflammatory response by MAPK and AKT/NF-κB signaling pathways.


INTRODUCTION
Mastitis, a common clinical disease in both animals and humans (1,2), is an acute condition with obvious symptoms of inflammation. Uncontrolled mastitis can be extremely harmful to lactating women (3,4). Moreover, mastitis has a potential negative effect on breastfeeding and is a main cause of considerable physical and psychological pain during lactation (5,6). In general, antibiotic treatment is the most commonly used strategy for treating acute mastitis. However, while antibiotics generally have a positive effect on bacteriostasis (7), they are not useful for controlling the development of subsequent inflammation (8).
Licochalcone A is a phenolic chalcone compound initially isolated from Glycyrrhiza uralensis in 1975 by Saitoh (9). The structure of licochalcone A and various biological activities have been established to date (Figure 1) (10), including antiinflammatory (11), anti-oxidation (12), antiseptic, and antitumor properties (13,14). As a traditional Chinese medicine monomer with anti-inflammatory and anti-bacterial properties (15,16), it has several advantages over traditional antibiotics. Drug resistance to this compound is relatively uncommon (17), along with a significant inhibitory effect on inflammation. Therefore, we speculate that licochalcone A may also have a protective effect on mastitis.
The blood-milk barrier is important for mammary gland resistance to the external environment (18). Destruction of the blood-milk barrier aggravates bacterial infection and promotes the development of inflammation (19). Acute mastitis is usually caused by infection of gram-negative bacteria and explosive proliferation (20). Escherichia coli, a common gram-negative bacterium, is widely present in the external environment and an important cause of acute mastitis (21). Lipopolysaccharide (LPS) in E. coli can cause severe immune response to mammary gland (22). However, LPS also destroys the blood-milk barrier and promotes the occurrence and development of inflammation (23). So in our experiments, mouse mastitis model was constructed using LPS extracted from E. coli. LPS can activate MAPK and AKT/NF-κB signaling pathways, which causes mouse mammary epithelial cells (mMECs) to release a large number of proinflammatory factors, destroy the epithelial barrier and accelerate the development of mastitis (24,25). Therefore, inhibiting the release of pro-inflammatory factors in mMECs and protecting the tight junction of mMECs may play an important role in alleviating mastitis. So the purpose of our study was to explore the effect of licochalcone A on inflammatory responses and bloodmilk barrier in LPS-induced in vivo and in vitro mastitis models.
Cell Culture mMECs purchased from American Type Culture Collection (ATCC R CRL-3063 TM ) were cultured in DMEM containing 10% FBS at 37 • C in a humidified incubator under 5% CO 2 .

Animal Experiments
BALB/c mice (8∼10 weeks old, 25∼30 g weight) were purchased from the Center of Experimental Animals of Baiqiuen Medical College of Jilin University (Jilin, China). Animals were housed in certified, standard laboratory cages and administered food and water ad libitum before experimental use. All animals care and experimental procedures were conducted in accordance with the guidelines established by the Jilin University Institutional Animal Care and Use Committee (approved on 27 February 2015, Protocol No. 2015047), designated "Guide for the Care and Use of Laboratory Animals" and approved by the Institutional Animal Care and use Committee of the University of Arizona. Animal studies were performed in compliance with arrive guidelines (26,27).

Animal Experimental Design
Lactating mice (5∼7 days after birth of offspring) were randomly divided into five groups: control (n = 10), LPS (0.2 mg/mL, 50 µL) treatment (n = 10), LPS + licochalcone A (5 mg/kg) (n = 10), LPS + licochalcone A (10 mg/kg) (n = 10), and LPS + licochalcone A (15 mg/kg) (n = 10). On day seven of lactation, the control group was injected with normal saline and the three treatment groups were injected different concentrations of licochalcone A (5, 10, and 15 mg/kg). After 1 h, LPS was injected into the fourth inguinal mammary gland of mice. After 12 h, the licochalcone A treatment group was injected with the same dose of different concentrations of licochalcone A. At 24 h after LPS injection, mice were anesthetized with pentobarbital sodium (45 mg/kg) and the mammary glands were collected.

Evaluation of Histological Changes
For histological analysis of the mammary gland, mice were euthanized and the four pairs of mammary glands fixed in 4% paraformaldehyde, followed by dehydration with ethanol. After paraffin embedding, 5 µm sections were cut and stained with hematoxylin and eosin (H&E) according to a previously described protocol (28). H&E-stained sections were examined under a light microscope to evaluate pathological changes. Simultaneously, a standard assessment method was conducted to determine mammary gland injury. Overall mammary gland injury was scored based on edema, neutrophil infiltration and hemorrhage, and three visual fields observed for each slice. Studies were performed in a blinded manner. Injury scores were representative of severity (0, no damage; 1, mild damage; 2, moderate damage; 3, severe damage; 4, very severe damage).
ELISA for TNF-α, IL-6, and IL-1β The protein levels of TNF-α, IL-6, and IL-1β in mammary glands were determined by ELISA kits according to the manufacturer's instructions.

FITC-Albumin Treatment to Evaluate Alveolar Tight Junction Permeability
To visualize alveolar tight junction permeability, untreated mammary glands, or those injected with 50 µL (0.2 mg/mL) LPS and 5, 10, or 15 mg/kg licochalcone A + LPS, were treated with FITC-conjugated albumin according to a previously described protocol (29). In brief, mice were deeply anesthetized with pentobarbital and the fourth mammary gland surgically exposed. The mammary gland was immersed in 2 mL PBS containing 3 mg/mL FITC-albumin to expose the interstitial side of alveolar epithelial cells for 10 min. After treatment with FITCalbumin, the mammary gland was washed in PBS three times and immersed in PBS containing 4% paraformaldehyde for 15 min. The pre-fixed mammary gland was embedded in optimal cutting temperature (OCT) compound and frozen in liquid nitrogen after which 5 µm cryosections were obtained. Cryosections were stained with 4' ,6-diamidino-2-phenylindole (DAPI) and images observed under a fluorescence microscope (TCS SP5; Leica, Mannheim, Germany).

Immunofluorescence Assays
As above, the mammary gland was embedded in OCT compound and frozen in liquid nitrogen to acquire 5 µm cryosections. Cryosections of mammary glands were fixed with 4% paraformaldehyde in PBS (pH 7.4) for 40 min at room temperature, and then incubated with PBS containing 5% donkey serum albumin to block non-specific interactions. After blocking, the sections were treated with primary antibodies diluted in 5% donkey serum albumin solution overnight at 4 • C, and then washed with PBS three times for 5 min each. After washed with PBS, the sections were exposed to secondary antibodies diluted in 5% donkey serum albumin solution for 1 h at room temperature. Controls were treated in a similar manner, except that primary antibodies were excluded. Finally, all sections were washed three times with PBS and stained with DAPI for 5 min. Images of stained sections were obtained under a fluorescence microscope.

qRT-PCR Analysis
Total RNA was isolated from cultured mMECs with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and amplification reactions conducted to detect gene levels of TNF-α, IL-6, and IL-1β. The primer sequences are shown in Table 1.

Item
Primer Length (bp)

Statistical Analysis
Images were generated using GraphPad prism software (La Jolla, CA, USA). Animals were randomly assigned to groups. In mouse studies, histological analysis was conducted in a blind design.
In cases where overall F-tests were significant (P < 0.05), post hoc comparisons using Tukey's method of adjustment were conducted to determine the location of significant pairwise differences. Analyses were performed using GraphPad prism 7.00 software.

Licochalcone A Administration Alleviates Pathological Injury of Mammary Gland in LPS-Induced Mice Mastitis
To study the protective effect of licochalcone A on mastitis, the mammary glands were collected, sectioned and stained with H&E ( Figure 2). There were no pathological injuries of mammary gland from control group mice (Figure 2A). Compared with control group, mammary gland from LPS group presented severe pathological injuries, such as thickening of the alveolar wall, inflammatory cell infiltration ( Figure 2B). However, these pathological injuries were alleviated by Licochalcone A in a dosedependent manner (Figures 2C-F, Supplementary Figure 3).

Licochalcone A Inhibits MPO Activity in LPS-Induced Mice Mastitis
Neutrophils are the first line of innate immune defense against infection. The accumulation of neutrophil in mammary gland Frontiers in Immunology | www.frontiersin.org   reflected the level of inflammation. In order to detect the degree of inflammation, the activity of MPO, which is a marker of neutrophil, was detected. The result found that LPS injection significantly increased MPO activity in mammary gland tissues compared with control group (Figure 3A). Compared with LPS group, licochalcone A treatment remarkably reduced the MPO activity in a dose-dependent manner ( Figure 3A).

Licochalcone A Reduces Pro-Inflammatory Mediators Production in LPS-Induced Mice Mastitis
Pro-inflammatory cytokines play important roles in the development of inflammation. To elucidate the effect of licochalcone A on the level of inflammation in LPS-induced mice mastitis, the production of the three major pro-inflammatory cytokines (including IL-6, IL-1β, and TNF-α) in mammary gland tissues were detected by ELISA. As shown in Figure 3, LPS significantly increased the protein levels of IL-6 ( Figure 3B), IL-1β (Figure 3C), and TNF-α (Figure 3D), and licochalcone A treatment suppressed this effect in mammary gland. iNOS and COX-2 are two important pro-inflammatory proteins correlated with inflammation. In our experiments, the protein levels of iNOS and COX-2 were also examined by western blot. The result found that LPS injection leading to remarkably production of iNOS (Figures 3E,F) and COX-2 ( Figures 3E,G), and licochalcone A administration significantly inhibits LPS-induced production of iNOS and COX-2 in mammary gland.

Licochalcone A Repairs the Blood-Milk Barrier Integrity in LPS-Induced Mice Mastitis
Blood-milk barrier, a specific structure of mammary gland, which could restrict the invasion of microbial. In order to elucidate the effect of licochalcone A on blood-milk barrier, the leakage of FITC-albumin from the interstitial side into alveolar lumen was determined. The result found that FITC-albumin was distributed in the interstitial side in the NT group, and FITC-albumin permeated into alveolar lumen in the LPS group ( Figure 6). As expect, licochalcone A could inhibit LPS-induced permeation of FITC-albumin (Figure 6).

Licochalcone A Enhances Protein Levels of Tight Junction Proteins and Restores Localization Changes of Claudin-3 and Occludin in LPS-Induced Mice Mastitis
Tight junction, the basic structure of blood-milk barrier, which could restrict the paracellular flow of aqueous molecules, ions, water, and microbes. To clarify the mechanism of licochalcone A on repairing the blood-milk barrier integrity, the protein levels tight junction components (including ZO-1, Occludin, and claudin3) and the localization changes of claudin-3 and occludin were determined by western blot and immunofluorescence assays. Protein levels of claudin-3, occludin and ZO-1 in the LPS group were significantly lower than those in the LPS + licochalcone A group (Figures 7A-D). Immunofluorescence staining revealed significantly lower fluorescence intensities in the LPS group, compared to the LPS + licochalcone A groups (Figure 7E), suggesting that licochalcone A repairs the tight junction destroyed by LPS. Weaker fluorescence intensities of claudin-3 and occludin were observed in the LPS group, compared to the control and LPS + licochalcone A groups. Claudin-3 localized with occludin at the apical-most regions in mammary epithelial cells (MECs) before LPS injection with no differences between localization patterns (Figure 8). At 24 h after LPS injection, claudin-3 mainly localized with occludin at the apical-most regions. Single localization of claudin-3 without occludin was also observed around these areas. In the presence of increasing doses of licochalcone A, claudin-3 re-localized with occludin at the apical-most regions, suggesting that licochalcone A increases the protein levels of tight junction and restores the original localization patterns of claudin-3 and occludin.

Licochalcone A Inhibits LPS-Induced Inflammatory Response in mMECs
In our in vivo experiments, we have proven that licochalcone A have anti-inflammatory role in LPS-induced mice mastitis.
To further this function of licochalcone A, the effect of licochalcone A on LPS-induced inflammatory response in mMECs was examined. Firstly, the cytotoxicity of licochalcone A, LPS and licochalcone A + LPS on mMECs was determined; The result found that licochalcone A, LPS and licochalcone A + LPS have no cytotoxicity on mMECs ( Figure 9A). Secondly, mMECs were pretreated with licochalcone A (1.2, 1.8, 2.4, and 3 µg/mL) for 1 h and then stimulated with LPS for 4 h, the mRNA levels of IL-6, IL-1β and TNFα were determined by qRT-PCR, and the protein levels of COX2 and iNOS were determined by western blot. The result of qRT-PCR shown that LPS significantly increased the mRNA levels of IL-6, IL-1β, and TNF-α compared with control and licochalcone A (3 µg/mL) treatment groups, and licochalcone A notably inhibited the increased mRNA levels of IL-6, IL-1β, and TNF-α (Figures 9B-D). The western result found that licochalcone A significantly inhibited LPS-induced production of COX2 and iNOS in a dose-dependent manner ( Figures 9E-G). These results are consistent with those in vivo, which further confirms the anti-inflammatory effect of licochalcone A.

Licochalcone A Inhibits LPS-Induced Activation of AKT/NF-κB and MAPK Signaling Pathways in mMECs
To further elucidate the anti-inflammatory mechanism, we also examined the effect of licochalcone A on LPS-induced activation of AKT/NF-κB and MAPK signaling pathways in mMECs. Firstly, mMECs were pretreated with licochalcone A (1.2, 1.8, and 2.4 µg/mL) for 1 h and then stimulated with LPS for 4 h, the phosphorylations of AKT, ERK1/2, JNK1/2, and P38 were determined by western blot. The result found that licochalcone A inhibited LPS-induced phosphorylations of AKT (Figures 10A,B), ERK1/2 (Figures 10A,E), JNK1/2 (Figures 10A,C), and P38 (Figures 10A,D) in a dose-dependent manner. Secondly, mMECs were pretreated with licochalcone A (2.4 µg/mL) for 1 h and then stimulated with LPS for 15, 30, 45, and 60 min, the phosphorylation of IκBα and NF-κB examined by western blot. With increasing time-periods, the NF-κB pathway was more strongly activated in the LPS group. Conversely, in the LPS + licochalcone A group, the NF-κB pathway was gradually suppressed, implying that inhibition of inflammation in mMECs by licochalcone A is time-dependent (Figures 10F-I).

Licochalcone A Enhances the Protein Levels of ZO-1, Occludin, and Claudin3 in mMECs
In our in vivo experiments, we have found that licochalcone A repaired blood-milk barrier integrity by increasing the protein levels of tight junction components (including ZO-1, Occludin, and claudin3). To further clarify whether licochalcone A increase the protein levels of tight junction components, mMECs were treated with 2.4 µg/mL for 24 h, and the protein levels of including ZO-1, Occludin, and claudin3 were detected by western blot. As shown in Figure 11, licochalcone A remarkably increased the protein levels of ZO-1 (Figures 11A-D), Occludin ( Figures 11A-D), and claudin3 (Figures 11A-D) in dosedependent manner.

DISCUSSION
Mastitis is a common clinical disease in humans and animals by microbial invasion. Antibiotic therapy is the current mainstream treatment but prone to drug resistance with significant sideeffects (30). Our findings demonstrated that licochalcone A could significantly alleviate mastitis via reducing inflammatory response and protecting the blood-milk barrier in LPS-induced mice mastitis. The further mechanistic study found that  Frontiers in Immunology | www.frontiersin.org licochalcone A significantly suppressed the MAPK and AKT/NF-κB signaling pathways in vivo and in vitro. These data suggested that licochalcone A might be a strong candidate for treatment of mastitis. Previous reports have shown that monomer composition could relieve mastitis, such as magnolol. Magnolol played an anti-inflammatory effect by inhibiting TLR4 (31). The traditional Chinese medicine monomer, licochalcone A, is reported to exert anti-inflammatory effects (32). Our team has found that licochalcone A could protect dopaminergic neurons via inhibiting neuro-inflammation in LPS-induced PD rat model (11). There is no report about the effect of licochalcone A on development of mastitis. Therefore, we examined the effect and the mechanism of licochalcone A on mastitis in vivo and in vitro. In vivo, we set up three concentration gradients, and found that the mastitis gradually relieved with the increase of licochalcone A concentration. It was also obvious from H&E staining that a large number of neutrophils were gathered in the mammary acini, and the thickening of the acinus was enhanced. As expected, licochalcone A could improve these micropathological changes. In addition, we found that licochalcone A also significantly inhibited the protein or gene levels of iNOS, COX-2, IL-6, IL-1β, and TNF-α in vivo and in vitro. At the same time, licochalcone A can also alleviate the inflammatory response of macrophages (Supplementary Figure 2). These results indicate that licochalcone A can significantly alleviate mouse mastitis.
Previous results have confirmed the anti-inflammatory effect of licochalcone A, but its anti-inflammatory mechanism has not been studied, so we studied the anti-inflammatory mechanism of licochalcone A in subsequent experiments. Interestingly, we found that licochalcone A can inhibit the phosphorylation of MAPK and AKT/NF-κB signaling pathways. Previous studies indicate that LPS triggers TLR4 downstream inflammatory pathways to stimulate the release of inflammatory cytokines (33). Two of the most important signaling pathways in inflammation are TLR4/MYD88/MAPK (Supplementary Figure 1) and AKT/NF-κB (34). The MAPK signaling pathway plays a key role in regulation of gene expression and cytoplasmic function (35). ERK1/2, JNK1/2, and P38 proteins in the MAPK signaling pathway are involved in the inflammation and stress response of the body and present crucial anti-inflammatory drug targets (36,37). The AKT/NF-κB signaling pathway mainly activates NF-κB by phosphorylating AKT to degrade IκBα (38). However, in the MAPK signaling pathway, p-ERK1/2 and p-P38 directly activate the NF-κB signaling pathway to promote inflammation (39). In vivo, with a gradual increase in the licochalcone A dose, protein levels of p-ERK1/2, p-JNK1/2, and p-P38 in MAPK signaling pathway were decreased. Our in vitro data were consistent with those obtained in vivo. However, the protein level of ERK1/2 was additionally inhibited in vivo. Therefore, licochalcone A reduced phosphorylation of ERK1/2 by inhibiting ERK1/2 protein levels. The protein levels of ERK1/2 did not change in vitro, which might be due to the short stimulation time, but its phosphorylation level was also inhibited. In subsequent experiments, we found that licochalcone A also inhibited the activation of AKT/NF-κB signaling pathway. In vivo and in vitro results showed that p-AKT, p-IκBα and p-P65 levels in the LPS group were significantly increased. However, licochalcone A concentration-dependently inhibited the protein levels of p-AKT, p-IκBα, and p-P65 in vivo and in vitro. Our experiments collectively indicate that licochalcone A inhibited the expression of inflammatory cytokines via inhibiting MAPK and AKT/NF-κB signaling pathways, thereby alleviating mastitis.
Previous studies have shown that blood-milk barrier is an important barrier for mammary gland resistance to invasion by external bacteria (40). Mastitis increases blood-milk barrier permeability and disruption of the tight junction, which can further aggravate the condition (41). Claudin-3, occluding, and ZO-1 are important tight junction proteins in MECs and their destruction directly affects the tight junction between cells (42). LPS is reported to damage the tight junctions between MECs (23). In our experiment, we have studied the anti-mastitis mechanism of licochalcone A. But how did licochalcone A affect the blood-milk barrier? Interestingly, we found that FITCalbumin in LPS group could enter in mammary gland acinus, and FITC-albumin in mammary gland acinus was decreased gradually with the increase of licochalcone A dosage in vivo. We also found that licochalcone A could increase the protein levels of claudin-3, occludin, and ZO-1 in vivo and in vitro. These data suggested that licochalcone A improved the integrity of bloodmilk barrier via increasing the protein levels of tight junction proteins. These data suggested that licochalcone A alleviated mastitis at least partially by improving the integrity of blood-milk barrier. This suggests that licochalcone A may play an important role in protecting the blood-milk barrier.
All our experiments indicate that licochalcone A reduces the expression of inflammatory factors by inhibiting the MAPK and AKT/NF-κB signaling pathway, and repairs blood-milk barrier damage caused by inflammation, supporting a novel clinical application of this compound in the treatment of mastitis.

AUTHOR CONTRIBUTIONS
YY, XK, QG, YL, YC, JW, DX, and HM helped with the experiments. SF, BL, and JL revised the content of the article. All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication. (A-C) Total protein in mMECs was collected after 4 h of LPS stimulation. Licochalcone A was added 1 h before LPS stimulation. Protein levels of TLR4 and MYD88 were detected via western blot and quantitatively assessed via densitometry using β-tubulin as an internal control. Protein levels were measured using ImageJ software (http://imagej.nih.gov/ij/) and normalized to that of β-tubulin. Values are presented as means ± SD (n = 3) ( * p < 0.05, * * p < 0.01, * * * p < 0.001, and * * * * p < 0.0001 vs. LPS group).