LncRNA XIST mediates bovine mammary epithelial cell inflammatory response via NF‐κB/NLRP3 inflammasome pathway

Abstract Objectives The correlations between long non‐coding RNAs (lncRNAs) and diverse mammal diseases have been clarified by many researches, but the cognition about bovine mastitis‐related lncRNAs remains limited. This study aimed to investigate the potential role of lncRNA X‐inactive specific transcript (XIST) in the inflammatory response of bovine mammary epithelial cells. Materials and methods Two inflammatory bovine mammary alveolar cell‐T (MAC‐T) models were established by infecting the cells with Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus). The expressions of pro‐inflammatory cytokines were measured, and the proliferation, viability and apoptosis of the inflammatory cells were evaluated after XIST was knocked down by an siRNA. The relationship among XIST, NF‐κB pathway and NOD‐like receptor protein 3 (NLRP3) inflammasome was investigated using an inhibitor of NF‐κB signal pathway. Results The expression of XIST was abnormally increased in bovine mastitic tissues and inflammatory MAC‐T cells. Silencing of XIST significantly increased the expression of E. coli or S. aureus‐induced pro‐inflammatory cytokines. Additionally, knockdown of XIST could inhibit cell proliferation, suppress cell viability and promote cell apoptosis under inflammatory conditions. Furthermore, XIST inhibited E. coli or S. aureus‐induced NF‐κB phosphorylation and the production of NLRP3 inflammasome. Conclusions The expression of XIST was promoted by activated NF‐κB pathway and, in turn, XIST generated a negative feedback loop to regulate NF‐κB/NLRP3 inflammasome pathway for mediating the process of inflammation.

damage, cell death, cytokine production, 5 changes in cell proliferation and migration, 6 etc Several signalling pathways are involved in inflammatory reaction, and nuclear factor-κB (NF-κB) is considered as a vital one.
Activated NF-κB contributes to regulating the process of cell proliferation and apoptosis, and the production of inflammatory cytokines including tumour necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6 and IL-8. 7,8 In addition, current studies have shown that activated NF-κB pathway could function as an upstream activator of the NOD-like receptor (NLR) family member pyrin domain-containing protein 3 (NLRP3), which is known as the inflammasome consisted of NLRP3, ASC and serine protease caspase-1. Inflammasomes are cytoplasmic multiprotein complexes, including four individual inflammasome branches (NLRP1, NLRP3, NALP4 and AIM2). 9,10 Inflammasomes can be generated and activated by various stimulators, 11 and E. coli and S. aureus identified in bovine mastitis are considered as main stimuli. 12,13 A few studies have reported that the complex network of NF-κB and NLRP3 inflammasome played an important role in mastitis, 14 but the specific mechanism between NF-κB and NLRP3 inflammasome needs to be further identified.
Long non-coding RNAs (LncRNAs) are defined as a novel class of transcripts with more than 200 nucleotides in length that generally lack the capability of coding protein and that were once regarded as transcriptional noises. 15 Increasing evidences have indicated that ln-cRNAs exert crucial functions in gene regulation, biological process and several diseases, such as inflammation 16 and immune response. 17 LncRNA X inactivate-specific transcript (XIST), a 17-kb-long RNA transcribed by the inactive X chromosome, is thought to be involved in the X chromosome inactivation in female mammals, inducing the compensation for X-linked gene dosage imbalance between the sexes. 18,19 More recently, accumulating evidences suggested that the expression of XIST was up-regulated in diverse cancer types including breast cancer, 20 gastric cancer, 21 lung cancer 22 and cystic fibrosis, 23 involving in cell proliferation, migration, invasion as well as apoptosis. 24 As the microenvironment of tumour is similar to the inflammatory condition, 25 XIST may also play a potential role in inflammatory process.
This study aimed to investigate whether XIST played a critical role in regulating the inflammatory response to pathogenic stimulus through NF-κB pathway and further identify the relationship between NF-κB pathway and NLRP3 inflammasome in bovine mastitis.

| Tissue specimens, cell line culture and cell transfection
Normal and mastitic mammary tissues used in this study were stored at −80°C in our laboratory before use, and the tissue collection procedure was described in our previous publication. 26 The bo-

| Stimulation of MAC-T cells with inactivated Escherichia coli and Staphylococcus aureus
Escherichia coli strain ATCC29213 and S. aureus strain ATCC25922 were obtained from the American Type Culture Collection (Manassas, VA). Bacteria were grown in Luria-Bertain (LB) broth at 37°C overnight, and the bacterial suspensions were washed by phosphate-buffered saline (PBS) for 3 times. Next, the bacteria were diluted in PBS to a series of concentration gradients. For each dilution, 10 µL was put onto LB agar medium overnight. The density of bacterial suspension was calculated by the colony count. Meanwhile, E. coli and S. aureus cells were killed by heating for 1 hour at 65°C and 2 hours at 60°C, respectively. Successful inactivation was verified when no bacteria colony was observed after overnight incubation on LB at 37°C.
The MAC-T cells were seeded into 6-well plates overnight at 37°C. Then, the cells were stimulated with inactive E. coli or S. aureus. The ratio of MAC-T cells to E. coli and S. aureus was 1:1000 and 1:100, respectively. After 24 hours of stimulation, the MAC-T cells were harvested for subsequent analyses.

| RNA extraction and Real-time quantitative PCR
The total RNA from the mammary tissues or cells was extracted using TriZol solution (TransGene, Shanghai, China) according to the manufacturer's instructions. The assessment of the quantity and quality of the total RNA was conducted by a spectrophotometer (NanoDrop Technologies, Wilmington, DE). The first-strand cDNA was generated from 3 µg total RNA using TransScript II First-Strand cDNA Synthesis SuperMix (TransGene). Real-time quantitative PCR (RT-qPCR) was carried out to examine the levels of genes by iQ5 light cycler (Bio-Rad, Hercules, CA) in 20 µL reactions. Finally, the expression of each gene was normalized to GAPDH. The primers used are listed in Table S1.

| Western blot analysis
The MAC-T cells were washed 3 times with PBS for 5 minutes and

| Cell viability assay
Cell viability was performed using Cell Counting Kit-8 (TransGene).
Briefly, cells were plated in 96-well plates at the same density of

| Flow cytometry
The percentage of apoptosis cells was assessed by Flow cytometry analysis with a commercial apoptosis assay kit (TransGene). Cells were collected and washed three times by cold PBS and then were double strained with Annexin V/PI according to the manufacture's instructions. After staining, the cells were analysed by flow cytometer. Finally, viable cells, early stage apoptotic cells, late stage apoptotic cells and death cells were determined with WinMDI software.

| Enzyme-linked immunosorbent assay
The protein secretion level of TNF-α, IL-6, IL-1β and IL-18 in each group was measured by enzyme-linked immunosorbent assay

| Statistical analysis
The statistical data were analysed with the Statistical Product and Service Solutions (SPSS 19.0 for Windows; IBM). One-way ANOVA test was used to determine the differences. Unless otherwise specified, the data were expressed as mean values ±standard deviation (SD). *P < 0.05 and **P < 0.01 was considered statistically significant. Each treatment was repeated at least three times.

| LncRNA XIST was up-regulated in bovine mastitic tissues and in inflammatory bovine mammary epithelial cell models
To explore the potential function of XIST during the process of bovine mastitis, we firstly performed RT-qPCR analysis to measure the expression level of lncRNA XIST in bovine mammary tissues. We found lncRNA XIST was significantly up-regulated in bovine mastitic tissues compared with counterpart normal tissues ( Figure S1 and Figure 1).
In addition, two inflammatory MAC-T cell models of bovine mastitis were established by infecting the cells with heat-killed E. coli and S. aureus, because these two heat-inactivated bacteria could induce the similar immune responses as live E. coli and S. aureus. 27,28 The expression levels of three pro-inflammatory cytokines (TNF-α, IL-6 and IL-1β) were significantly up-regulated (Figure 2A), indicating a successful establishment of bovine mastitis cell model. Consistent with the expression trend in bovine mastitic tissues, lncRNA XIST expression level was also significantly higher in inflammatory MAC-T cells than that in normal MAC-T cells ( Figure 2B). These results indicated that increased expression of XIST was critically involved in the process of bovine mastitis.

| Knockdown of XIST intensified Escherichia coli or Staphylococcus aureus-induced inflammatory response in MAC-T cells
Considering that XIST was up-regulated in MAC-T cells in response to E. coli or S. aureus infection, we wondered whether XIST was in-

| XIST mediated cell proliferation, viability and apoptosis of inflammatory MAC-T cells
The cell proliferation was evaluated by EdU incorporation assay.
The results showed the number of EdU positive cells in E. coli or S. aureus-treated group was lower than that in control group, in-

| XIST suppressed the activation of NF-κB signalling pathway and reduced the production of NLRP3 inflammasome
As XIST could regulate the expression of pro-inflammatory cytokines, we speculated that XIST might be involved in the activation of upstream NF-κB pathway, a well-known inflammation-related signal pathway, and its target genes. To confirm our speculation, we firstly performed dual-luciferase reporter assays to examine the activation of NF-κB. Results showed that knockdown of XIST significantly enhanced the activity of the NF-κB luciferase reporter gene under inflammatory conditions ( Figure 5A). In addition, we found the phosphorylations of p65 and IκB had a significant increase in the inflammatory MAC-T cells transfected with Si-XIST compared with those transfected with Si-NC ( Figure 5B). Moreover, immunofluorescence staining showed that knockdown of XIST could promote the translocation of NF-κB p65 subunit from cytoplasm to nucleus in E. coli or S. aureus-induced inflammatory MAC-T cells ( Figure 5C).
These results suggest that E. coli or S. aureus-triggered activation of NF-κB pathway was suppressed by XIST.
The NLRP3 inflammasome has been demonstrated to play a crucial role in the process of inflammation of mammary glands. 29 To elucidate whether the anti-inflammatory mechanism of XIST was associated with NLRP3 inflmmasome, we further examined the levels of NLRP3 inflmmasome components, including NLRP3,

pro-casepase-1 and ASC, in inflammatory MAC-T cells. The results
showed that E. coli or S. aureus was able to induce the formation of NLRP3 inflammasome, as indicated by the increased mRNA F I G U R E 3 The expression levels of pro-inflammatory cytokines in XIST knockdown cells. A, An siRNA was designed for XIST (Si-XIST) gene loci and the knockdown efficiency was detected by RT-qPCR. A non-targeting siRNA (Si-NC) was used as control. B and C, Gene and protein expression levels of TNF-ɑ, IL-1β and IL-6 in Escherichia coli or Staphylococcus aureus-induced inflammatory MAC-T cells transfected with Si-NC or Si-XIST were analysed by RT-qPCR and ELISA, respectively. All data were presented as the mean ± SD (n = 3). *P < 0.05, **P < 0.01  Figures 3C and 6C). These results suggested that XIST was able to inhibit E. coli or S. aureus-induced production of NLRP3 inflammasome.

| XIST mediated inflammatory response through the NF-κB/NLRP3 inflammasome signalling pathway in MAC-T cells
The above data showed that silencing of XIST could activate NF-κB pathway and enhance the production of NLRP3 inflammasome in E. coli or S. aureus-treated MAC-T cells. We further explored the relationship among XIST, NF-κB and NLRP3 inflammasome using an inhibitor of NF-κB signal pathway, BAY 11-7082. The data showed F I G U R E 5 The effect of XIST on NF-κB signal pathway. A, The MAC-T cells were co-transfected with siRNA, NF-κB luciferase plasmids and Renilla luciferase plasmids, and then stimulated with Escherichia coli or Staphylococcus aureus. The luciferase activity of the NF-κB reporter was measured by Dual-Luciferase Reporter Assay System and the activity of firefly luciferase was normalized to that of Renilla luciferase. B, The activations of two key signal subunits of the NF-κB pathway (p65 and IκB) were analysed by Western blot and subsequent quantitative analysis. GAPDH was used as a loading control. C, Immunofluorescent staining was conducted to show the localization of NF-κB p65 subunit (Green). The nuclei were coloured with DAPI (blue). All data were presented as the mean ± SD (n = 3). *P < 0.05. Abbreviations: Si-NC, a negative control siRNA; Si-XIST, an siRNA targeting bovine XIST; -EC and -SA, MAC-T cells were stimulated with Escherichia coli (E coli) and Staphylococcus aureus (S aureus); pp65, phosphorylated p65; pIκB, phosphorylations of IκB that BAY 11-7082 significantly blocked E. coli or S. aureus-induced phosphorylation of NF-κB p65 in inflammatory MAC-T cells transfected with Si-XIST ( Figure 7A). In addition, the increased expressions of NLRP3, ASC and pro-caspase-1 after infection with E. coli or S. aureus was obstructed in the presence of BAY 11-7082 ( Figure 7B).
The increased secretion levels of NLRP3 inflammasome end products (IL-1β and IL-18) induced by E. coli or S. aureus were also blocked in XIST knockdown cells ( Figure 7C). Furthermore, inhibiting the activation of the NF-κB pathway also blocked E. coli or S. aureus-induced up-regulation of XIST in MAC-T cells (Figure 8). These date suggested that the production of NLRP3 inflammasome and the transcription activation of XIST triggered by E. coli or S. aureus were the consequences of the activation of NF-κB signalling pathway.

| D ISCUSS I ON
Gram-negative E. coli and gram-positive S. aureus are two of the most common pathogens that often result in different types of bovine mastitis during both lactation and non-lactating periods. Previous studies have described that E. coli and S. aureus could evoke a cascade of immune responses via being recognized by TLRs. 30 Besides, heat-inactivated E. coli could increase the expression of both TLR2 and TLR4, while heat-inactivated S. aureus could only promote the expression of TLR2 in bovine mammary epithelial cells. 31 Subsequently, TLRs signal transduction pathways trigger signalling cascades in the early innate immune response and eventually activate NF-κB factors, 32 as well as the NLRP3 inflammasome. 33 Successfully activated NF-κB augments the production of downstream pro-inflammatory cytokines such as pro-IL-1β, IL-6 and TNF-α through binding to their promoter regions. 34 Then, the activated molecular platform NLRP3 is responsible for the formation of the active secreted protein proteolytic caspase-1, which cleaves pro-IL-1β into mature IL-1β. 35 Therefore, signalling, we investigated whether the intracellular signalling pathways of NF-κB and NLRP3 inflammasome were regulated by XIST, and we found that silencing of XIST could enhance E. coli or S. aureus-induced NF-κB phosphorylation, as well as the activation of NLRP3. It is well known that NLRP3 inflammasome functions as an inflammatory factor in generating inflammation by maturating IL-1β and IL-18. 43 In the present study, knockdown of XIST increased the expression of NLRP3 inflammasome components (NLRP3, ASC and pro-caspase-1) under inflammatory conditions. Then, we further found knockdown of XIST could promote NLRP3 inflammasome end products IL-1β and IL-18 secretion. These results suggested that E. coli or S. aureus-induced activation of NLRP3 inflammasome was inhibited by XIST. In addition, NLRP3 inflammasome is involved in the process of NF-κB-mediated inflammation. 44 We further found that the blockade of NF-κB pathway could reduce NLRP3 inflammasome component expression under the inflammatory condition.
Therefore, these results suggested that the function of XIST in the process of E. coli or S. aureus-induced mastitis was to prevent the production of inflammatory cytokines, particularly, through inhibiting the activation of NF-κB. Besides, we also demonstrated that the production of NLRP3 inflammasome was based on the activation of NF-κB pathway in the inflammatory bovine mammary epithelial cells.
Interestingly, our data showed that inhibition of NF-κB pathway could abate E. coli or S. aureus-induced up-regulation of XIST in MAC-T cells, suggesting that increased XIST expression was associated with NF-κB pathway. It is reported that nuclear lncRNAs usually have the function of regulating transcription by influencing transcription factors activity. 45 XIST was mainly located in nuclear, so it might regulate transcription in the same way as nuclear lncRNAs. Based on our data, we concluded that the interaction of XIST and NF-κB might generate a nuclear RNA-protein complex and F I G U R E 8 Escherichia coli or Staphylococcus aureus-induced XIST transcription depended on NF-κB pathway. The expression levels of XIST in MAC-T cells were measured by RT-qPCR under inflammatory conditions with or without the NF-κB inhibitor (BAY 11-7082). All data were presented as the mean ± SD (n = 3). *P < 0.05 vs control, NS means no significance among all columns F I G U R E 9 A proposed model depicting the possible mechanism of XIST on mediating cell proliferation, viability and apoptosis via generating a negative feedback regulation of NF-κB/NLRP3 inflammasome pathway. The recognition of Escherichia coli or Staphylococcus aureus by Toll-like receptors (TLRS) triggered the activation of NF-κB signalling, thus stimulating the production of inflammatory genes such as TNF-α, pro-IL-1β, IL-6 and NLRP3 inflammasome. These inflammatory genes constituted an inflammatory microenviroment to modulate cell proliferation, viability and apoptosis. In turn, NF-κB-dependent XIST interacted with NF-κB signal pathway and inhibited the overexpression of downstream inflammatory genes the NF-κB-controlled downstream genes were then regulated by it ( Figure 9).
In conclusion, we have shown for the first time that lncRNA XIST expression was up-regulated in bovine mastitic tissues and inflammatory mammary epithelial cells induced by bacteria in vitro. XIST promoted cell proliferation, maintained cell viability and inhibited cell apoptosis under inflammatory conditions through inhibiting the activation of NF-κB pathway and the formations of pro-inflammatory cytokines. Furthermore, activated NF-κB pathway could promote the expressions of both XIST and NLRP3 inflammasome, and in turn, XIST generated a negative feedback loop to regulate the NF-κB/NLRP3 inflammasome pathway for mediating the process of inflammation. Bovine mastitis is a localized defence response of the body components against the pathogens, but excessive and sustained inflammation leads to cell and tissue damage. So, the function of XIST in bovine mastitis was to inhibit this excessive and sustained inflammatory process. Understanding the key role of lncRNA XIST in bovine mammary epithelial cells under inflammatory conditions will contribute to the identification of new therapeutic targets to treat bovine mastitis.

ACK N OWLED G EM ENTS
We thank Prof. Mark D. Hanigan (Virginia Polytechnic Institute and State University, Blacksburg, VA) for his gift of the MAC-T cell line.

CO N FLI C T O F I NTE R E S T S
The authors declare no conflict of interest.

AUTH O R CO NTR I B UTI O N S
Ma and Gao performed study design. Ma and Pei performed data collection. Gao and Zhang involved in contribution of new reagents or analytical tools. Ma, Pei, Wang and Feng performed data analysis.
Ma, Pei and Gao involved in manuscript preparation.